report 5.4 decision support tool...design4energy!d5.4 decision support tool development page 5 of 64...

Design4Energy - Building life-cycle evolutionary Design methodology able to create Energy-efficient Buildings flexibly connected with the neighborhood energy system

Co-funded by the European Commission within the 7th Framework Programme. Grant Agreement no: 609380. 2013-10-01…2017-09-30 (48 months).

Report 5.4 Decision support tool

Revision ......................................... 1 Preparation date ............................ 2016-06-14 (m13) Due date ........................................ 2017-06-30 (m45) Lead contractor ............................. FHR

Authors: Team Fraunhofer .......................... Fraunhofer IAO Farid Fouchal ................................ Loughborough University Vanda Dimitriou ............................. Loughborough University Tarek Hassan ................................ Loughborough University Steven Firth ................................... Loughborough University Argyris Oraiopoulos ....................... Loughborough University

Design4Energy ! D5.4 Decision support tool development of 64

2017-07-23

Disclaimer The information in this document is provided as is and no guarantee or warranty is given that the information is fit for any particular purpose. The user thereof uses the information at its sole risk and liability. The document reflects only the author’s views and the Community is not liable for any use that may be made of the information contained therein.


2017-07-23

Table of contents

1. Executive Summary ............................................................................................... 7

2. Introduction ............................................................................................................. 82.1 Purpose and target group ......................................................................................... 82.2 Contributions of partners ........................................................................................... 82.3 Baseline .................................................................................................................... 82.4 Relations to other activities ..................................................................................... 10

3. State of the Art for the decision support tool .................................................... 113.1 Background ............................................................................................................. 113.2 Existing methods and techniques for decision support in retrofit and maintenance 123.3 Existing methods and techniques in decision support for retrofit and maintenance of buildings on neighbourhood scale ................................................................................. 173.4 Life cycle analysis and cost benefit analysis ........................................................... 19

4. Concept of the D4E decision support tool ......................................................... 204.1 D4E decision support workflow ............................................................................... 204.2 Architecture of D4E decision support tool ............................................................... 22

5. Retrofit and maintenance alternatives generator .............................................. 265.1 Workflow of Alternatives Generation ....................................................................... 265.2 Creating retrofit or maintenance alternative combinations ...................................... 29

6. Targets setting ...................................................................................................... 326.1 Requirements .......................................................................................................... 326.2 Key performance indicators weighting .................................................................... 326.3 Benchmark database .............................................................................................. 336.4 Key performance indicators benchmarking ............................................................. 35

7. KPIs for retrofit alternatives ................................................................................ 377.1 Calculation of KPIs for each alternative .................................................................. 377.2 Checking and filtering out alternatives using benchmarks ...................................... 387.3 Import of KPI weighting ........................................................................................... 39

8. Ranking process of alternatives ......................................................................... 408.1 Topsis method ......................................................................................................... 408.2 Ranking process and algorithms ............................................................................. 41

9. Decision Support Tool GUI .................................................................................. 439.1 Background on GUI for DST outcome .................................................................... 439.2 Development of prototype for visualisation of design alternatives KPIs .................. 479.3 Validation of the DST GUI prototype with end users ............................................... 48

10. Implementation of the retrofit and maintenance DST in real world case study .......................................................................................................................... 49

10.1 GSM building Revit model ..................................................................................... 4910.2 Generation of alternatives ..................................................................................... 50


2017-07-23

10.1 Calculating KPIs for retrofit alternatives ................................................................ 5110.2 Filtering of retrofit alternatives ............................................................................... 5310.3 Ranking alternatives and presentation of results .................................................. 54

11. Conclusions ........................................................................................................ 5711.1 Summary of achievements .................................................................................... 5711.2 Relation to continued developments ..................................................................... 5711.3 Other conclusions and lessons learned ................................................................ 57

12. Acronyms and terms .......................................................................................... 58

13. References .......................................................................................................... 60


2017-07-23

Table of figures Figure 1 Multi criteria decision making (MCDM) process being adopted in D4E .................. 11

Figure 2 D4E Retrofitting Decision Support Workflow ........................................................... 21Figure 3 Operation of the building including retrofit or maintenance ...................................... 22

Figure 4 System architecture of the decision support tool and the supporting components .... 25Figure 5 Workflow of alternatives containing the options generator and evaluation & decision making. ...................................................................................................................................... 26Figure 6 Management of users and projects ............................................................................. 27

Figure 7 Dashboard in the user-interface .................................................................................. 27Figure 8 Selection of Components from the Component Library/ Choosing the alternative components / Collaborative Database Concept ........................................................................ 28Figure 9 Automatic Generation of new gbXML-components. ................................................. 29

Figure 10 Generation of alternatives based on the chosen components presented as gbXML files. Download and provision of a high amount of generated alternatives as a zip file. ......... 29

Figure 11 User-defined weights within the Target Setting Tool of the D4E platform ............. 33Figure 12 Building benchmark overview ................................................................................. 34

Figure 13 Description of benchmark buildings ........................................................................ 35Figure 14 KPI target values ...................................................................................................... 36

Figure 15 The EplusKPI Tool for calculation of retrofit alternatives KPIs, the input and output files ............................................................................................................................................ 37

Figure 16 Checking and filtering out alternatives using benchmarks in the D4E platform ..... 38Figure 17 User-defined weighting as exported from the Target Setting Tool of the D4E platform ..................................................................................................................................... 39Figure 18 The ranking tool, the input files and the tool output ................................................ 41

Figure 19 Ranking tool algorithm in Visual Studio using VB.NET ......................................... 42Figure 20 Graphical user interface of the proposed model (part1) ........................................... 43

Figure 21 Graphical user interface of the proposed model (part 2) .......................................... 44Figure 22 REFIT Energy Consumption Report ........................................................................ 45

Figure 23 A snapshot in time of traditional and emerging design practices[66]. ..................... 46Figure 24 Four levels of creativity [66] .................................................................................... 46

Figure 25 Three forms of energy use visualisation tested by the ENLITEN project [67] ........ 47Figure 26 Prototype GUI of the DST KPIs ............................................................................... 48

Figure 27 The GSM case study ................................................................................................. 50Figure 28 Generation of alternatives in the D4E alternatives generator ................................... 51

Figure 29 Using the alternative gbXML files to calculate the KPIs ......................................... 52Figure 30 The calculated KPIs in the kpiXML and short kpiXML files .................................. 52


2017-07-23

Figure 31 User defined weights for ranking ............................................................................. 54

Figure 32 The magnitude of closeness to the ideal solutions used for ranking of the retrofit alternatives using TOPSIS ........................................................................................................ 55

Figure 33 Display of KPI results for three best alternatives using the prototype UI ................ 56


2017-07-23

1. EXECUTIVE SUMMARY The objective of task 5.4 is the development of the decision support tool. It will involve: (i) the development of prototype tools to automatically generate maintenance/ retrofit advice at different levels of abstraction and for use by different stakeholders, (ii) incorporating ICT visualisation and user-centred design techniques; (iii) create strong engagement with the various stakeholders (including industry, consumer and policy) to refine the key features of the retrofit decision support tool and the delivery mechanisms for retrofit or maintenance advice. Report 5.4 is therefore structured into an analysis of the current state and background information especially in the area of decision support tools for energy efficient maintenance and retrofitting of districts and buildings and the detailed description of the components of the decision support tool developed in the D4E project. This includes the retrofit and maintenance alternatives generator, the target setting for key performance indicators, the decision support tool for ranking design alternatives and the graphical user interface. Finally, the decision support tool is implemented in a real-world case-study.

For the development in the D4E project, existing solutions are used either as a basis or for learning based on previous applications or experiences. The architecture of the D4E decision support tool includes the process phases of data collection and specification, both on relevant technologies, components and information on benchmarking buildings as well as case specific information on relevant buildings to provide information for different stakeholders and integrate decision support into early phases of the maintenance and retrofitting value chain. Within the alternative generator, options are integrated into a gbXML model that can directly be applied in a major part of common simulation tools able to calculate energy efficiency of buildings. The aim of providing a package on best available solutions for energy efficient design of buildings and districts based on the individual ranking of available options is thereby supported by the automated generation of alternatives, the energy performance analysis tool, individual criteria options and finally on ranked options for energy efficient maintenance or retrofitting.


2017-07-23

2. INTRODUCTION

2.1 Purpose and target group The objective of task 5.4 is the development of the decision support tool. It will involve: (i) the development of prototype tools to automatically generate maintenance/ retrofit advice at different levels of abstraction and for use by different stakeholders, (ii) incorporating ICT visualisation and user-centred design techniques; (iii) create strong engagement with the various stakeholders (including industry, consumer and policy) to refine the key features of the retrofit decision support tool and the delivery mechanisms for retrofit or maintenance advice. The decision support tool is thus principally based on the analysis and evaluation of potential combinations of different technologies (options) that enable all stakeholders to better understand the underlying performance of components and technologies and to improve decision making on these potential alternatives. Target groups concerned by the decision making process are all building or district stakeholder that are concerned by the building or district throughout the life-cycle. This includes planners, designers, general contractors, engineers as well as building users and stakeholders involved in maintenance and repair or refurbishment of buildings and districts. The increased level of transparency that is generated through the decision support tool enables an increased integration of these stakeholders along the value chain as well as an improved efficiency and effectiveness of decision making in early phases of the building and district life-cycle.

2.2 Contributions of partners Using the complementary results of the design4energy, D4E project as a basis, the decision support tool that is in fact supported by other tools has being created jointly by LU and FHR. The options generator on technologies and components in this task has been principally developed by the Fraunhofer Institute for Industrial Engineering (FHR) and Loughborough University (LU). LU has developed the optimisation part which included a set of tools and processes to enable decision making, including the ranking process. FHR has set up a collaborative system based on the semantic media wiki to collect and share information on technologies and components and has succeeded to integrate this information into the building and district value chain through the usage of building data, its combination with component and technology data to automatically generate options for building and district simulation models. LU has worked on the application of benchmarking data to evaluate design alternatives to each other and thus identify the best available alternative.

2.3 Baseline The efforts for effective improvement of the overall energy performance of buildings led to development of numerous methodologies. They all proceed by initial energy audits for estimation of the building’s energy status and for proposing then evaluation of different scenarios regarding the upgrade to a more energy efficient building. Among the early methodologies that followed this process are: ‘‘Tobus’’, ‘‘Xenios’’ and ‘‘Epiqr–Investimno’’, introduced through the JouleII, Altener and Growth programmes [2]. For achieving the objectives set out by the EPBD directives, innovative tools were also developed, such as the ‘‘EPA-ED’’ and ‘‘EPA-NR’’, for assistance in energy audits and support the designers and experts in buildings’ energy certification [3]. A ‘‘Datamine’’ database has also been created


2017-07-23

for collecting and storing data from the buildings’ energy certification in Europe [4]. Furthermore, computerisation has gradually been introduced into construction industry causing radical changes in the way buildings are conceptualised, designed, built, operated, maintained and retrofitted [5]. More specifically building information modelling (BIM) is leading the change at every stage of the building’s life cycle orchestrated by a growing ICT community of practice [6]. The digital environment enabled by BIM allows numerous dynamic activities by multi-disciplinary stakeholders, including collaboration, information sharing and data storage throughout the building lifecycle [9]. Gero et al. [7] were among the first to introduce a multi-criteria model in building design for enabling trade-offs between the building energy (thermal) performance and other criteria such as capital cost and usable area. Similar approaches have since been adopted by more scientists ([8][10][11][12]). Kaklauskas et al.[13] developed a multivariate design method and multi-criteria analysis for retrofit underperforming buildings, by calculating the significance, priorities and utility degree of various alternatives and ranking them. Allane [14] used a multi-criteria knapsack model to select the best retrofit option at the concept stage of the project. Juan et al. [65] developed a self-learning driven decision support system for assessing housing condition and propose best retrofit actions taking into account trade-off between cost and quality. Although the described approaches appear to have enabled design and selection of good enough retrofit actions they are also limited as they are applied upon a set of predefined and pre-evaluated alternative solutions which are not holistic and exhaustive [15]. Juan [65] investigated how multi objective optimization could support in solving the problem of the improvement of the energy efficiency of existing buildings by maximising the possible number of alternative solutions and energy efficiency measures. Juan [65] showed that no optimal solution exists for the problem due to the competitiveness of the involved decision criteria. Given the small number of considered solutions combinations versus all possible combinations there is no guarantee that the utmost best solution is considered at all. On the other hand, when a large number of solutions are considered the evaluation and selection process becomes technically not feasible. Multi-objective optimization is a scientific area that offers a wide variety of methods with great potential for the solution of complicated decision problems. Asadi et al. [16] used simulation-based multi-objective optimization scheme (a combination of TRNSYS, GenOpt and a Tchebycheff optimization technique developed in MATLAB) to optimize the retrofit cost, energy savings and thermal comfort of a residential building. Alternative materials for walls insulation, roof insulation, window types, and installation of a solar collector were considered. As an evolutionary multi objective algorithms which is necessary for tackling the problem and considers approximation methodologies like regression modelling of the building in the optimization for wide and holistic account of all possible combination to achieve the best retrofit option. Ibn-Mohammed et al. [17] integrated economic considerations with operational and embodied emissions into a decision support system for the optimal ranking of building retrofit options to effectively manage the reduction of lifecycle environmental impacts. He integrated economic and net environmental benefits and focuses on marginal abatement cost methods and Pareto optimisation. The work undertaken in this Task aligns with the work conducted in two previous EU projects which are ISIS and HESMOS, however in this task we have used a gbXML approach while in the two others they have followed an IFC based approach.


2017-07-23

2.4 Relations to other activities The resulting data from T5.3 analysis, T5.2 specification and T5.1 the bidirectional communication protocol were used as an input for the Design4Energy platform to optimise the design and hence be able to generate different scenarios for optimised design for retrofitting and devise maintenance action. This includes the development of prototype tools, the incorporation of ICT development for user-centred design and finally the creation of strong stakeholder engagement focussing on stakeholders involved in decision making on retrofitting solutions in early design stages. The Target setting tool developed by VTT (WP2) was used to obtain the weights and benchmarks used in the decision making process and to align with the Key Performance Indicators used in the retrofit scenario. GSM provided the case-study building (WP8) used for development and demonstration of the tools being created for Decision Making.


2017-07-23

3. STATE OF THE ART FOR THE DECISION SUPPORT TOOL

3.1 Background Many energy efficiency design platforms have already been developed. The value of design platforms is in their workflow speed and quality, facilitating team contribution integration, and rapid feedback on design aspects such energy performance [47]. OpenStudio is an open source project to create a collection of software tools for energy modelling, daylight analysis and various other simulations. In OpenStudio the community of developers can program plugins, applications, analysis tools to support all kind of stakeholders to design more energy efficient building [46]. For these platforms, databases are essential for their functioning. While xBIM is another open source development platform, which allows creating application for BIM based on the IFC standard. The corresponding development library contains far more functions to manipulate the IFC files. The libraries can be integrated in a .net environment and are mostly written in C#. The currently supported IFC version by xBIM is IFC2x3 [48]. TNO BIM Server is an open source development platform, which allows creating application for BIM based on the IFC standard. The BIMserver allows mainly querying, merging and filtering the BIM-model and generating IFC files on the fly. Further important functions include versioning, notifications, geo-locating models, authentication and plug-in infrastructure. The libraries can be used in a Java environment. The currently supported IFC version by TNO BIM is IFC2x3 [49]. The BuildingSMART Data Dictionary (bSDD) is a reference library or a framework that aims at supporting improved interoperability in the building and construction industry. It can connect software applications to product databases or attach specific attributes to construction designs. These references can include information from a product manufacturer, typical room requirements, cost data or environmental data [50][51].

Multiple Criteria Decision Making (MCDM) integrates multiple indicators into a single meaningful index to allow ranking and comparing options for decision making. It is an efficient statistical method to combine component indices arising from all the information sources into a single overall meaningful index, therefore ranking and comparing are feasible. MCDM has the ability to weight different alternatives and make judgement on various criteria for possible selection of the best/suitable alternative(s). A typical MCDM problem is when there are a number of criteria to assess a list of alternatives. Each alternative is represented by a single value for each of the criteria to permit the assessment and/or ranking, see Figure 2. Complex decision requires consideration of multiple criteria [29].

Figure 1 Multi criteria decision making (MCDM) process being adopted in D4E


2017-07-23

The weighted sum model (WSM) is the simplest of MCDM, which is applicable only when all the data are expressed in exactly the same unit otherwise the final result, is ironic. The weighted product model (WPM) can be considered as a modification of the WSM, and has been proposed in order to overcome some of the WSM weakness. The analytic hierarchy process (AHP), as proposed by Saaty [37] is a later development and it has recently become popular. Some other widely used methods are the ELECTRE and the TOPSIS methods. Hwang [21] have developed a number of MCDM techniques to meaningfully integrate many indices to an overall index in order to decide on the ranking of a number of alternatives. They have developed an MCDM approach called “Technique for order preference by similarity to ideal solution” (TOPSIS) [21]. Filar [36] described in detail the TOPSIS method and used entropy as a basis to determine the importance of weights and applied the MCDM technique to assess the possibilities available. An MCDM process include: (i) determining the relevant criteria and alternatives; (ii) attach numerical measures to the relative importance of the criteria and the impact of the alternatives on these criteria and (iii) process the numerical values to determine a ranking of each alternative. If compared, WPM uses multiplication of these numerical values in the model instead of addition in WSM. However, the Analytic Hierarchy Process (AHP) method that is based on priority theory, decomposes a complex multi-dimensional decision making problem into a system of hierarchies. It uses the relative importance of the alternatives in terms of each criterion. The AHP has the ability to logically incorporate data and expert’s judgement in the model for measurement and prioritising intangibles. As a complex and unstructured situation is broken down, its components are arranged into a hierarchic order including criteria and alternatives. The core of MCDM consists of the construction of pairwise comparison matrices and the extraction of weights by means of the principal right eigenvector. Linguistic variables in a fuzzy environment in the form of Triangular Fuzzy Numbers (TFNs) are also used to determine the weights of criterion.

3.2 Existing methods and techniques for decision support in retrofit and maintenance

Many decision support tools have been developed for architects and building designers to help them choose the best building design options with retrofit and maintenance in mind. The following deals with analysis of existing decision support tools in the context of energy efficient buildings design concerning their current capabilities and the fulfilment of the identified requirements.

3.2.1 Energy retrofit intelligent decision support system (ERIDSS) Using an earlier developed information model, an energy retrofit intelligent decision support system (ERIDSS) was developed which integrates expert knowledge with quantitative information to provide homeowners with accurate information for decision-making [30].

There are three major components: the knowledge-based management, data management, and the user interface subsystems. A key component of the ERIDSS is the knowledge-based management subsystem (KMS) which provides the heuristic/implicit knowledge elicited from the energy retrofit experts and is complemented by quantitative information from the data management subsystem in order to assist users with energy retrofit decision making [30]. Earlier work by the involved researchers led to the development of an energy retrofit decision process (ERDP) model. This model consisted of three main parts: (1) identify retrofit measures, (2) shortlist and prioritize measures, and (3) provide expert advice on installation. Based on this model, the determinants of energy retrofit expert knowledge and the elicitation of such knowledge; the following knowledge-based modules in the KMS for the proposed IDSS were developed [30]:


2017-07-23

1. Knowledge-based thermal envelope measures (KB-Therm) module. 2. Knowledge-based heating, cooling, and air conditioning measures (KB-HVAC)

module. 3. Knowledge-based hot water heating measures (KB-HWH) module. 4. Knowledge-based lighting measures (KB-Light) module. 5. Knowledge-based stand-alone and energy saving measures (KB-SES) module. 6. Knowledge-based expert advice on installation (KB-EAI) module.

3.2.2 TRNSYS TRNSYS is used by a number of researchers. Florides et al. [32] used TRNSYS to examine measures such as natural and controlled ventilation, solar shading, various types of glazing, orientation, shape of buildings, and thermal mass aiming to reduce the thermal load, while Zurigat et al. [33] used TRNSYS to evaluate different passive measures aiming to reduce the peak cooling load of school buildings [31].

3.2.3 EnergyPlus EnergyPlus is used by Becker et al. [34] to assess specific factors of building design elements (window orientation, glazing type, thermal resistance of walls, etc.) and 20 ventilation strategies for schools’ energy consumption and efficiency [31].

3.2.4 Visual DOE Visual DOE is used by Tavares and Martins [35] to perform a sensitivity analysis that results in energy efficient design solutions for a specific case study. A significant number of predefined solutions are modelled and evaluated [31].

3.2.5 Multiple criteria decision making (MCDM) MCDM integrates multiple indicators into a single meaningful index to allow ranking and comparing options for decision making. It is an efficient statistical method to combine component indices arising from all the information sources into a single overall meaningful index, therefore ranking and comparing are feasible. MCDM has the ability to weight different alternatives and make judgement on various criteria for possible selection of the best/suitable alternative(s). A typical MCDM problem is when there are a number of criteria to assess a list of alternatives. Each alternative is represented by a single value for each of the criteria to permit the assessment and/or ranking. Complex decision requires consideration of multiple criteria [29].

An MCDM process include: (i) determining the relevant criteria and alternatives; (ii) attach numerical measures to the relative importance of the criteria and the impact of the alternatives on these criteria and (iii) process the numerical values to determine a ranking of each alternative [36].

The core of MCDM consists of the construction of pairwise comparison matrices and the extraction of weights by means of the principal right eigenvector. Linguistic variables in a fuzzy environment in the form of Triangular Fuzzy Numbers (TFNs) are also used to determine the weights of criterion [36].

3.2.6 The weighted sum model (WSM) WSM is the simplest of MCDM, which is applicable only when all the data are expressed in exactly the same unit otherwise the final result, is ironic [36].

3.2.7 The weighted product model (WPM)


2017-07-23

WPM can be considered as a modification of the WSM, and has been proposed in order to overcome some of the WSM weakness [36]. WPM uses multiplication of numerical values in the model instead of addition in WSM [36].

3.2.8 The analytic hierarchy process (AHP) AHP, as proposed by Saaty [37] is a later development and it has recently become popular [36]. AHP method that is based on priority theory decomposes a complex multi-dimensional decision making problem into a system of hierarchies. It uses the relative importance of the alternatives in terms of each criterion. The AHP has the ability to logically incorporate data and expert’s judgement in the model for measurement and prioritising intangibles. As a complex and unstructured situation is broken down, its components are arranged into a hierarchic order including criteria and alternatives [36].

3.2.9 ELECTRE ELECTRE is a multi-criteria method to evaluate the thermal comfort, acoustic comfort and indoor air distribution in an office air-conditioned room. The purpose of the procedure is to extract design rules for air-conditioning systems that satisfy indoor comfort requirements [31].

3.2.10 ELECTRE III ELECTRE III method is used to rank the three strategies discussed in chapter “Alternative Options and Strategies” after the ascending and descending ranking. Each criterion is weighted according to its importance. The retrofitting strategies are placed in a final ranking if they have the same position in the ascending and descending ranking [31].

3.2.11 Office rating methodology (ORME) ORME was proposed by Roulet et al. [38] and it uses ELECTRE algorithms to rank office buildings based on comfort, waste and energy consumption criteria. The ORME method introduces the energy efficient retrofit score (EERS). Different predefined scenarios are evaluated using the EERS [31].

3.2.12 Effective retrofitting of an existing building Tool (REFLEX) Pasanisi and Ojalvo [39] developed an MCDA tool called REFLEX for building refurbishments. The multi-criteria approach ranks the retained solutions according to the end users’ point of view and energy suppliers’ satisfaction [31].

3.2.13 Energy performance indoor environmental quality retrofit method for apartment building refurbishment (EPIQR)

EPIQR is used in the literature for energy efficiency and indoor environment using multi-criteria decision aid [31].

3.2.14 Tool for selecting office building upgrading solutions (TOBUS) TOBUS is used in the literature for energy efficiency and indoor environment using multi-criteria decision aid for office buildings [31].

3.2.15 BEMS


2017-07-23

In the operational phase of a building, energy efficiency is achieved by data collection and BEMS. BEMS can contribute to a significant reduction of the energy consumption of buildings and improvement of the indoor comfort through advanced control techniques [40]. Modern control systems provide an optimized operation of the energy systems while satisfying indoor comfort. A comparison of the various control schemes for buildings (Proportional-Integral-Derivative, On-Off, fuzzy, etc.) is provided by Kolokotsa [41]. Recent technological developments based on artificial intelligence techniques (neural networks, fuzzy logic, Gas, etc.) offer several advantages compared with classical control systems. A review of the fuzzy logic contribution in indoor comfort regulation as well as HVAC control and energy efficiency is performed by Kolokotsa [42]. The role of neural networks in buildings is analysed by Kalogirou [43]. A model-based supervisory control strategy for online control and operation of a building is presented by Ma et al. [44][31].

3.2.16 ECOSOFT ECOSOFT (IBO – Austria) is an assessment software of the environmental impact during construction and maintenance [45].

3.2.17 Technique for order preference by similarity to ideal solution (TOPSIS) TOPSIS is a widely used method. Hwang [21] has developed a number of MCDM techniques to meaningfully integrate many indices to an overall index in order to decide on the ranking of a number of alternatives. They have developed an MCDM approach called TOPSIS. Filar [36] described in detail the TOPSIS method and used entropy as a basis to determine the importance of weights and applied the MCDM technique to assess the possibilities available.

3.2.18 OpenStudio OpenStudio is an open source project to create a collection of software tools for energy modelling, daylight analysis and various other simulations. In OpenStudio the community of developers can program plugins, applications and analysis tools to support all kind of stakeholders to design more energy efficient building [46]. The value of design platforms is in their workflow speed and quality, facilitating team contribution integration, and rapid feedback on energy performance [47].

3.2.19 xBIM xBIM is another open source development platform, which allows creating application for BIM based on the IFC standard. The corresponding development library contains far more functions to manipulate the IFC files. The libraries can be integrated in a .net environment and are mostly written in C#. The currently supported IFC version is IFC2×3 [48].

3.2.20 TNO BIM server TNO BIM Server is an open source development platform, which allows to create application for BIM based on the IFC standard. The BIMserver allows mainly querying, merging and filtering the BIM-model and generating IFC files on the fly. Further important functions include versioning, notifications, geo-locating models, authentication and plug-in infrastructure. The libraries can be used in a Java environment. The currently supported IFC version is IFC2×3 [49].

3.2.21 The BuildingSMART Data Dictionary (bSDD)


2017-07-23

bSDD is a reference library or a framework that aims at supporting improved interoperability in the building and construction industry. It can connect software applications to product databases or attach specific attributes to construction designs. These references can include information from a product manufacture, typical room requirements, cost data or environmental data [50][51].

3.2.22 The building component library (BCL) by NREL BCL by NREL provides searchable information about EE related technologies and a list of measures to meet energetic issues [52]. The included information can represent physical characteristics of buildings such as windows, walls and doors, or can refer to related operational information such as occupancy, equipment schedules and weather information. Each measure and energy system can be downloaded as a XML, RB and OSM file describing these components [53].

3.2.23 Data repository ISES Data Repository ISES is a cloud-based data repository. It contains information such as climate data or stochastic templates but most interestingly energy product and material catalogues containing energy properties of products and materials. The elements that are saved in this database are „ifcBuilding Element“, „ifcMaterial“, „ifcMaterialProperties sub-types“, and „ifcDistributionElement“. They are used for creating physical constructs incorporated into the building out of product and material properties [54]. It aims to bring component manufacturers effectively together with their customers in order to keep the market open for their products. Further it aims at choosing appropriate components by specialists. The library uses the PLIB ontology model (based on ISO 13584). All information is saved in the ifc file format [55].

3.2.24 The MagiCAD product database The MagiCAD Product Database is a product catalogue or database that contains over one million products from over hundred manufacturers. A designer can choose components through a plugin directly via the CAD-tool interface. This interface is connected to a plugin on the manufacturers’ site [56]. Other proprietary component databases are linked to CAD tools of component libraries with parametric objects provided by the software producer itself or by its developer community. The objects are stored in the CAD specific format, e.g. ArchiCAD library parts, and describe building elements as 2D CAD symbols, 3D models and text specifications for use in drawings, presentations and calculations [57].

3.2.25 The open energy information platform (OpenEI) OpenEI is a Media Wiki giving the most current information needed to make informed decisions on energy, market investment and technology development [58].

3.2.26 The NBS national BIM library for BIM users The NBS National BIM Library for BIM users is a collection of both generic and manufacturer objects which enables the use of BIM objects throughout a project. These contain an extensive number of generic objects that can be used for outlining design stages [36].

3.2.27 Eurobau The portal eurobau.com provides a data pool of construction materials. It was created for the implementation of the EU directive 89/106/EWG. The data content is generally structured by


2017-07-23

the application case of the materials, e.g. underground engineering, insulation or interior construction [59].

3.2.28 Masea Masea is a database accessible through a web platform which contains a set of different materials (such as painting, plaster, stones, insulations, wall cladding and others). The description of the materials contains detailed information about physical and structural parameters, pictures, hints on usage, diagrams with information among others about moisture retention, water absorption, drying process, and further information about manufacturer. The data of each material is downloadable as XML file. Further the database is used in the tools WUFI®, DELPHIN®, and EPASS HELENA [60][61].

3.2.29 National residential efficiency measures database by NREL National Residential Efficiency Measures Database by NREL provides a national unified database of residential building retrofit measures and associated costs [62].

3.2.30 Wiki of the climate-smart planning platform Wiki of the Climate-Smart Planning Platform aims at establishing a development community for openly accessible climate planning and modelling tools [63].

3.2.31 Building catalogue EMPORIS Building catalogue EMPORIS collects information on buildings (public and business). It sets information standards for facilitating and connecting the collected data [64].

3.3 Existing methods and techniques in decision support for retrofit and maintenance of buildings on neighbourhood scale

Over the past years, the need to implement energy and planning actions on an urban scale has given significant attention to modelling urban developments. The intermediate scale between individual buildings and the city as a whole is represented by the neighbourhood level that embraces many of the most influencing factors of carbon emissions. The neighbourhood scale is the immediate environment which surrounds a building and the occurring local conditions affect not only its performance but also its longevity. In that perspective, a number of decision support tools have been developed to investigate the relationships between the urban environment and the energy demand of buildings, allowing the identification of both energy intensive areas and influencing factors. This section presents the main existing decision support tools in the context of energy efficiency in the neighbourhood scale.

3.3.1 LT Urban This is a technique that combines the Lighting and Thermal tool with a Digital Elevation Model (DEM)[68]. The DEM provides morphological urban parameters which are then passed to the LT simulation engine to calculate the energy consumption of individual buildings. This allows the investigation of the impact of the urban texture on building energy consumption [69].

3.3.2 Energy and Environmental Prediction (EEP) The Energy and Environmental Prediction (EEP) model is an auditing tool that aims in aiding in the sustainable development of a city [70]. The EEP comprises of four sub-models; a domestic stock sub-model, using 100 building archetypes, a non-domestic stock sub-model, using 48 archetypes, an industrial sub-model for 16 different sectors and a traffic flow sub


2017-07-23

model. The energy use and CO2 emissions of the archetypes are estimated based on data from historical records and surveys and then the results are extrapolated to the entire building stock.

3.3.3 Energy Hub The Energy Hub is a model for optimising the energy efficiency in a neighbourhood [71]. The mathematical model of the energy hub can be combined with optimisation techniques depending on the design objectives. This allows the exploration of efficient configurations such as the optimal combination of renewable energy sources to reduce carbon emissions and the optimal retrofitting of existing buildings in neighbourhoods.

3.3.4 The City Energy Analyst The City Energy Analyst is a model for analysing and optimising building energy systems in neighbourhoods and districts [72]. It is based on seven different databases (including weather, georeferenced information of buildings, local topography, building stock archetypes, time series of occupancy and ventilation rates) and six different modules (including energy demand of buildings, supply networks, and other)[73]. The framework allows the analysis of energy, carbon, and financial benefits from urban design scenarios. It is a holistic approach to evaluating design and engineering solutions to assist urban planning authorities in increasing the energy efficiency of neighbourhoods and city districts.

3.3.5 CitySim CitySim is software that supports the sustainable planning of urban neighbourhoods. It integrates four models, a thermal model, a radiation model, a behavioural model and a plant and equipment model [74]. CitySim describes the factors affecting the demand and supply of energy, water and waste in a quick and efficient way, by simulating the urban microclimate, the human behaviour, the synergies between buildings and resources. It has been validated using the BESTTEST validation procedure [75]. It has been applied to model the effects of the UHI and the long–wave radiation exchange on the energy demand of buildings [76][77][78].

3.3.6 SimStadt SimStadt is a dynamic physical energy model of buildings, for simulating the energy demand of cities that enables direct parameter transfer between the 3D urban geometry data models and the energy related parameters required for building simulation [79].

3.3.7 CityBES The City Building Energy Saver (CityBES) is a web-based data and computing platform that focuses on the energy analysis and modelling of a city’s building stock for improved large scale efficiency programs. This platform is making use of an international open data standard to represent and exchange information, the CityGML, and the software programme EnergyPlus to simulate building energy use and savings from energy efficient retrofits [80].

3.3.8 CityGML CityGML is an open data model implemented as an application schema for the Geography Markup Language 3 (GML3), the extendible international standard for spatial data exchange issued by the Open Geospatial Consortium (OGC) and the ISO TC211 [81]. The aim of the development of CityGML was to “reach a common definition and understanding of the basic entities, attributes and relations within a 3D city model” [82].CityGML has also the ability to be extended in order to include more detailed data, such as urban noise, tunnel, bridges, utility networks, to improve data exchange and tools’ interoperability in the Urban Energy and


2017-07-23

Information Modelling, by the so called Application Domain Extensions (ADE). In this view, an international consortium of urban energy simulation developers established the Energy ADE in order to store and manage data required for Building Energy Modelling [83]. By including an Energy ADE to a CityGML file, building related information such as energy performance certifications, the reflectance, transmittance and emissivity of windows, occupancy schedules, electrical appliances, heat pumps, boilers, thermal zones and energy systems can be used to perform detailed building energy modelling. The level of detail in a CityGML Energy ADE model can be such that allows for hourly (or even sub-hourly) simulations in dynamic thermal simulation software at an urban scale.

3.4 Life cycle analysis and cost benefit analysis Whereas cost-benefit analyses or commonly used for the evaluation of technologies in built environments, life cycle analyses are less common, mainly due to required cost and effort. For the evaluation of technologies and components in the D4E decision support, elements from both are applied to enable the identification of best available solutions. This includes reference information on the cost of technologies and/or components, a qualitative description of potential benefits described in text form as well as information on the costs/benefits along the life-cycle of the component and/or technology.


2017-07-23

4. CONCEPT OF THE D4E DECISION SUPPORT TOOL The following chapters frame the concept for the decision support tool developed in D4E project. It is based on the research for existing solutions, requirements and methods done in the previous chapter. The chapter is split into the description of the workflow and the architecture for implementation.

4.1 D4E decision support workflow In Figure 2 the workflow developed to support retrofit decision making is shown. A BIM to BEM xml-based process previously developed and presented in report 5.3 [18] enables the use of Building Information Modelling for energy performance assessment of existing buildings. In addition, a series of tools have been developed for retrofit Decision Making Support: the Options Generator linked to the D4E component database; the EplusKPI tool for the retrofit alternatives; the ranking tool; and the results viewer. The Options Generator uses the base-case model for the existing building (the ‘as-is’ case) and the retrofitting options contained within the D4E component database to create a number of possible retrofit combinations. The EplusKPI tool calculates the Key Performance Indicators for each of the retrofit alternative combinations through comparison with the results for the base-case building KPIs. The ranking tool is then used to rank the retrofit alternative options taking into consideration both the energy and cost analysis and the end-users preferences using weights and benchmarks for the KPIs defined through the D4E platform and the target setting tool. Finally, a results viewer is used to offer feedback to the end-user and visual support for Decision Making.


2017-07-23

Start

BuildREVITmodelusingappropriateguidelines

Conductbuildingsurveyandcollectoperationaldata

Goodenergyperformance?

END

YESNO

Tool/processdevelopedKeys:

gbXMLEnhancerTool

gbXMLtoidfConversionTool

EplusKPIToolforbase-caseandretrofitalternatives

RankingTool

RunanalysisusingEnergy+

Optionsgenerator D4EComponent

DB

gbXML

Base-casegbXML

Base-caseandalternativegbXMLfiles

idffilesidffiles

csvfiles

EplusXMLfiles

ResultsevaluationResultsevaluation

InstructionsXML

IdfXML_template

Targetbenchmarks

XML

ReducedXML(ifrequired)ReducedXML(ifrequired)

WeightsXML

Resultsviewer

TargetSettingtool

Indexes

Filteringtool

D4EPlatformtools

Figure 2 D4E Retrofitting Decision Support Workflow

(Chapter5)

(Chapter7)

(Chapter6)

(Chapter9)

(Chapter8)


2017-07-23

4.2 Architecture of D4E decision support tool A detailed retrofit and maintenance scenario is first developed and described. The scenario starts with the facility manager evaluating the operation stage and maintenance data of an existing building and reveals some building performance changes which require serious attention such as undertaking some repair or upgrade to the building. An architect takes over and starts analysing historical data of operation, maintenance records as well as user behaviour data, monitoring data, the map of neighbourhood energy nodes and cost data. From this analysis, it becomes apparent that some of the data is not compatible with building’s energy anticipated performance. He/she therefore request a thorough investigation of the causes of the energy consumption mismatch with original design in specific parts of the building which involves the heating system.

A heating system expert is called in and identifies an old boiler as the source of the problem. The architect in collaboration with a building services engineer sketch a retrofit or maintenance design using a BIM model on the D4E platform. In doing so, the architect takes into consideration a number of parameters such as the local weather profile, facility management reports, financial status of the building owner and looks into other case studies to decide the best option forward for optimisation of energy level ahead of the conceptual design completion. At this stage, the architect considers the market and the various options for the energy performance of the project’s life cycle and cost of future operation and maintenance to prepare to discuss various design options with the client to make a decision. Mainly two routes become possible depending on the budget in hand and existence of information on new source of district heating to become available in the near future within the vicinity of the building. These options are analysed and evaluated by the designer comparing the retrofitting improvements versus maintenance action. The D4E platform supports the designers by highlighting critical building zones. The designer can filter out the existing building data for transferring them to certain design tools (CAD, etc.). Using the different design tools the designers can develop retrofitting variants for further integration and analysis in the collaborative platform. The simulation tool integrated in the system enables running a number of analyses to assess the impacts of the proposed retrofitting or maintenance variants on the energy efficiency and compare them to historical data of similar existing buildings. The design is then passed on to the mechanical and electrical engineers as a BIM model. The 3D collaborative environment provides them the possibility to explore what-if-scenarios, they can drag components from the database library to modify and optimise the design. Furthermore, the platform provides them with cost estimation of the different options on different terms (short to long). All the information required during these activities will be stored into a common database and accessed through the D4E platform. During the operation of the building the stages described in Figure 3 are followed:

Figure 3 Operation of the building including retrofit or maintenance


2017-07-23

The process of identifying building issues during the operation phase of the life cycle can be described in the following, where building under performance is identified and adequate measures are undertaken using D4E platform. Monitoring building operation (client & FM): During the operation stage of a building, the client (user/owner) and the facility management team undertake scheduled monitoring and/or observation of the building performance. The data being collected can be clustered under the following topics:

• Building survey • Sensors and monitoring data • Client report & user interview • Review of maintenance strategy • Access BIM files / As built drawings

Requirement definitions Based on operation data the facility manager defines their requirement to re-establish the normal operation of the building or to upgrade the performance level for example to comply with new regulations or simply respond to the client request. Data Analysis The energy expert and other energy system experts such as HVAC engineers will to assess the reported building energy performance issue and set targets for remedy or upgrade. The target setting will involve selection of key indicators and defining the operating ranges for each indicator. Review of project’s objectives The energy expert reviews the objectives traced to meet the use and operation needs. These are identified on the basis of the FM requirements and the expected energy performance of the building. Search benchmarks and finalise key target setting The benchmark browser and search tool will be used to set the benchmarking related targets. The standard methods and benchmarks for consideration in this project include CIBSE (Chartered Institute for Building Services Engineers) TM22, TM46, TM39, TM46 and TM 47, the AM11 Building Energy and Environmental Modelling (BEEM) (CIBSE Applications Manual 11) and the EPBD (Energy Performance of Buildings Directive), IPMVP (International Performance Measurement and Verification Protocol) in the USA, as well as the Performance Contracting Program standards ISO. Setting key target levels will be based on benchmarks and client requirements, which involve setting the range with its extreme boundaries for each indicator being considered in the project. Retrofit or maintenance option generation and selection Choosing of energy options, then run a feasibility study and produce a feasibility report for retrofit and maintenance will be carried out by the energy expert first through selecting variables to be used to form the options. Decision making by the client At this point of the process a review of alternatives will be undertaken by the client using the decision support tool to make an initial decision which is more suitable for the project, maintenance or retrofit. The feasibility results will be prepared in a format that will simplify the decision making of the client. Produce retrofit brief


2017-07-23

If the decision is to undertake retrofit, a model has to be created. It starts by the client producing the brief which contains expected energy performance, the KPIs and the targets being set. The next step is to execute the brief by first defining the specification of the ideal solution to the given client requirements. Creation of retrofit alternatives concept designs The remaining stages of the model will be completed by an Architect who will conduct an environmental analysis and building performance assessment. The architect generates a project program. Improve retrofit model The BIM model is then improved with material data for better energy efficiency performance and CO2 emissions reduction. Performance analysis for passive design The architect has the option to choose to undertake some building performance analysis under the proposed retrofit alternatives using the energy efficient BIM. This procedure will mainly help to reduce the number of options that can be considered and even provide enough information to make the ranking on the basis of the chosen criteria from the list of the performance indicators. Design concept and review by client The architect finalises design alternatives with KPIs profiles using BIM and generates a design concept for each potential alternative. Through the collaborative environment, the client reviews the produced design concepts taking into consideration of his/her main requirements which include energy consumption, the construction cost and LCC. Analysis of energy demand The energy expert will conduct an analysis of building energy demand using the energy simulation performance tool. Following this analysis, the client via the collaborative environment will review the energy options for the selected retrofit options and narrows down the number of options which would be passed onto the architect to verify the BIM models of the selected alternatives in term of their energy matching potential. Final approval of selected alternative by client The improved BIM models for the selected retrofit alternatives will be accessed through the collaborative environmental by the client for final approval. At this stage if there is more than one alternative they will be listed in a ranking order on the basis of the most important indicators to the client to enable fast approval. Detailed study of the final selected alternative and finalise the design ready for execution Through the collaborative environment, various experts that are relevant to the different components included in the final solution will be invited to access the BIM model which embeds the maintenance solution to verify that the solution selected is adequate to respond to the identified requirements.

In Design4Energy the process described above will be automated using a decision support tool that relies on data from an option generator and energy performance analysis tool that will later use a graphical user interface to display the output to the end users and the client. The following Figure 4 shows the system architecture for the construction and operation of the decision support tool. The decision support tool was structured using four building blocks which are: (i) energy performance and simulation block; (ii) retrofit and maintenance options generator; (iii) optimisation block; and (iv) the decision making block that is based on Multiple Criteria Decision Making (MCDM) method.


2017-07-23

Figure 4 System architecture of the decision support tool and the supporting components


2017-07-23

5. RETROFIT AND MAINTENANCE ALTERNATIVES GENERATOR The availability of alternatives at system level on selected retrofit and maintenance options for stakeholders in early design stages enables a more efficient and effective decision making. Within the following chapters, the user interface as well as the code used for the key functionalities in the alternatives generators is presented in more detail.

5.1 Workflow of Alternatives Generation The alternatives generator allows users to load building models into the system and get support for decisions on components through pre-configured building models containing selected alternatives. The workflow of the alternatives generator is shown in Figure 5. It start with a gbXML-file as an input. Its output can directly used for simulation of building performance of the selected retrofitting options.

Figure 5 Workflow of alternatives containing the options generator and evaluation & decision making.

The workflow involves two major aspects. First, is the options generator multi gbXML model. This allows users to feed an individual building model into the system and reflects the number of different building models that stakeholders are confronted with in building retrofitting. Second is the options generator in which components for retrofitting and maintenance are combined with the building model to process an output that can directly, without any further activity required from the user, be used as an input for simulation of building performance.

OptionsGenerator(OG)Project

LoadBuildingModel(s)

SelectComponentsbasedoncharacteristicsandKPIs

GenerateOptions

Redefineparameters(buildingorientation)

GenerateSimulationOutputs

DownloadmaingbXML

OutputSimulationTable

(„Report“-files as soon as uploaded at the SALD server)

gbXML-file based on the XML-standard


2017-07-23

To reflect the relation of single stakeholders to invidual retrofitting projects, the options generator allows the configuration of users and projects (see Figure 6) and incorporates a dashboard user interface (see Figure 7) where the user gets a direct overview on relevant options.

Figure 6 Management of users and projects

Figure 7 Dashboard in the user-interface

To enable the access to existing components, a wiki-based collaborative component library is used as a basis for the options generator.Components in this library are categorised according to trades or building component categories, e.g. roofs, doors, walls. The collaborative nature of the database shall reflect the value chain of the construction industry, where various stakeholders in project-based activities are continuously involved in decision making on components and technologies relevant to building and district retrofitting (see Figure 8).


2017-07-23

Figure 8 Selection of Components from the Component Library/ Choosing the alternative

components / Collaborative Database Concept Based on the fed-in gbXML files, combined with the alternatives chosen within the components library, the options generator produces gbXML file alternatives with the components included into the gbXML code. This allows the users to directly feed the resulting files into appropriate simulation models to get information on buidling behaviour (see Figure 9).


2017-07-23

Figure 9 Automatic Generation of new gbXML-components.

All files generated are available for download in a zip-file format or as single files (see Figure 10). This allows downloading of a large number of different alternatives in one single file for further usage in appropriate simulation models.

Figure 10 Generation of alternatives based on the chosen components presented as gbXML

files. Download and provision of a high amount of generated alternatives as a zip file. A key challenge to be adressed within dissemination of D4E results is the creation of a collaborative framework that enables the continuous update of the components database by different stakeholders. This is still considered as a major barrier for the long-term establishment of the D4E database in the construction industry, both for new construction projects as well as for retrofitting and maintenance of buildings and districts.

5.2 Creating retrofit or maintenance alternative combinations Within the following chapter, the logic of the programming code is presented, which allows the creation of retrofit and maintenance combinations. Each of the components is presented in a slightly different form within the gbXML file. Therefore, the logic varies accordingly and is presented in a different chapter. As an example, the code is shown only in the first part “5.2.1 Generating wall-elements”


2017-07-23

5.2.1 Generating wall-elements

1- Look for each Surface-node in the gbXML <Surface surfaceType="ExteriorWall" constructionIdRef="aim1068" exposedToSun="true" id="aim2814">

2- Get all constructionIdRef 3- Looking for the construction node with the id which equals constructionIdRef 4- Changing all the elements according to the alternative file

<Construction id="aim1068"> <U-value unit="WPerSquareMeterK">0.88893</U-value> <Absorptance unit="Fraction" type="ExtIR">0.8</Absorptance> <Roughness value="Rough" /> <LayerId layerIdRef="aim1072" /> <Name>Fibre cement wall</Name> </Construction>

5- From the Construction-node get the <LayerId layerIdRef="aim1072" /> 6- Look for the layer node with the same id

<Layer id="aim1072"> <MaterialId materialIdRef="aim0514" /> <MaterialId materialIdRef="aim0515" /> <MaterialId materialIdRef="aim0516" /> <MaterialId materialIdRef="aim0517" /> <MaterialId materialIdRef="aim0518" /> <MaterialId materialIdRef="aim0519" /> </Layer>

7- Change the layer node according to the layer node in the alternative file and keep the ID without change

8- Add all Material-nodes in the alternative file into the gbXML main building file

5.2.2 Generating roof-elements The gbXML file has more than one node of type <Surface surfaceType="Roof" … > and also these nodes reference to different constructionIdRef

For example, in the provided gbXML file, the file has - 114 nodes with <Surface surfaceType="ExteriorWall" … > - These nodes have three different constructionIdRef

The goal of the task is to exchange all roofs. There are roofs with three different construction IDs in the original gbXML. For each of these roofs (“Surface”-element) the following steps were applied:

1- Look for each Surface-node in the gbXML 2- Get all constructionIdRef 3- Looking for the construction node with the id which equals constructionIdRef 4- Changing all the elements according to the alternative file


2017-07-23

5- From the Construction-node get the LayerId 6- Look for the layer node with the same id 7- Change the layer node according to the layer node in the alternative file and keep the

ID without change 8- Add all Material-nodes in the alternative file into the gbXML main building file

5.2.3 Generating door-elements The gbXML file has more than one node of type <Opening constructionIdRef="aim1117" openingType="NonSlidingDoor"… >

Our task was to exchange all doors. For only one of these door (“Open”-element) following steps are applied:

1- Look for each ”Opening” node in the gbXML 2- Get its constructionIdRef 3- Looking for the construction node with the id which equals constructionIdRef 4- Changing all the elements according to the alternative file 5- From the Construction-node get the LayerId 6- Look for the layer node with the same id 7- Change the layer node according to the layer node in the alternative file and keep the

ID without change 8- Add all Material-nodes in the alternative file into the gbxml main building file

5.2.4 Generating window-elements The gbxml file has more than one node of type ”Opening” with attribute.

Our task was to exchange all Windows. For all of these windows (“Open”-element) following steps are applied.

1- Add a new window nodes in the alternative file into the gbxml main building file and

be sure that the added nodes have a unique identity id in order to be used later as a reference (windowTypeIdRef)

2- Look for each Open-node in the gbxml and change the windowTypeIdRef with new idRf of the new added window


2017-07-23

6. TARGETS SETTING An important part of the decision support process is the target setting. The tool and methodology use it to adapt the building model and choose the best alternative. These targets are called key performance indicators (KPI). Therefore, the first part deals with the description of requirements by the stakeholder. The second part discusses the impact or so-called weighting by the KPIs. The third shows how the KPIs can be stored and communicated through a centrally accessible and REST-ful web platform. The last part show how these indicators can be called and used for the decision making process.

6.1 Requirements Requirements for retrofitting and maintenance projects, both on a district and building level, highly depend on individual users or stakeholders. This leads to high-level volatility based on these preferences. This volatility has been reflected by a high-level flexibility within the options generator that is able to reflect individual requirements along the building or district life-cycle. General behaviour of components and technologies is described and evaluated in a general way within the descriptive area of the technology wiki. Furthermore, life-cycle behaviour is integrated into the gbXML model through selected component characteristics that are then combined with information available in the building model. This is due to the high level interdependency between building models and components in building or district simulation.

In addition to the integration of requirements along the design process through the options generator, performance indicators can be weighted according to individual user preferences. This is described in the following chapter.

6.2 Key performance indicators weighting The target of the Multi-Criteria Decision Making process for the retrofit scenario is set by the end-users, the architect and the client. The end-users' requirements are defined, quantified and stored using a designated graphical user interface (Figure 11). The targets are set on the basis of Key Performance Indicators (KPIs) weights and acceptable ranges definition. The end-user plays an integral role in the Decision Making process. The user's preferences are specified as weights of relative significance for each of the KPIs of the retrofit alternatives' energy and cost performance. The weights assigned to the KPIs are specified on a scale of 0-100. The higher the weight assigned the most significant the relevant KPI is for the user. A weight equal to 0 indicates complete dismissal of the impact on the related criteria whereas a weight of 100 signifies prioritisation of the specific KPI. The weights are normalised across the six main KPIs and are ready for use by the ranking tool. To guide the user during target setting, appropriate ranges of acceptable values are identified for each of the six KPIs. For the definition of the range limits appropriate benchmarks are being used [18]. The benchmarks relate to different levels of performance from A to E following the Energy Performance Certificate (EPC) rating scheme, with A and E signifying the best and least good performance. Based on the available information, the user can define the acceptable limits for the KPI values and, after the alternatives have been simulated, filtering out the options that do not meet their expectations is performed.


2017-07-23

Figure 11 User-defined weights within the Target Setting Tool of the D4E platform

Based on the user-defined weights and target values different alternatives of retrofitting and maintenance configurations can be compared with each other in order to support decisions on which components and technologies apply to districts and single buildings.

6.3 Benchmark database In addition to the options generator, which is the key aspect of the D4E decision support, the benchmarking database allows comparing the options with existing buildings that are used as a benchmark. Therefore, the benchmarking database contains information on buildings itself, as well as on the location and site of the building (see Figure 12).

This overview enables users to compare their retrofitting project with similar building types located in similar built environments and thus the selection of buildings which generate a value add for the decision making process.


2017-07-23

Figure 12 Building benchmark overview

The description of each single building in the benchmarking database provides characteristics in the three categories of the building itself, its location and the building site (see Figure 13). On the building level, this includes e.g. the facility management project phase of the building, the number of stories as well as the floor area, function and years of construction and operation of the building. The location of the building is specified through its address and regional location. The site is defined by the size of the overall site as well as the building efficiency and density.


2017-07-23

Figure 13 Description of benchmark buildings

The benchmarking database shall provide the user of the options generator with alternatives that have been built or refurbished in previous projects. The final comparison is done based on the D4E set of KPI target values that are described in more detail in the next chapter.

6.4 Key performance indicators benchmarking The key performance indicators are the key aspect of the building benchmark. They allow users to compare their options defined based on the options generator with existing buildings based on the building type, location and site. Key performance indicators are structured into the categories of (1) economic performance, (2) performance in use, (3) environmental performance and (4) impact on neighbourhood performance (see Figure 14).


2017-07-23

Figure 14 KPI target values

A more detailed calculation of the key performance indicators is done within the decision support tool of the D4E project and is described in the following chapters.


2017-07-23

7. KPIS FOR RETROFIT ALTERNATIVES The decision support tool is a procedure consisting of both IT-tools and an elaborated procedure based on recent research in building energy balances. The following chapter deals with the development of the decision support tool, after the retrofit alternatives have been generated, which includes the steps: calculation of KPIs, checking and filtering alternatives and importing the KPI weighting.

7.1 Calculation of KPIs for each alternative The next step of the decision making process is the calculation of the KPIs for each of the retrofit alternatives. A BIM to BEM xml-based process is used which relies on the gbXML BIM export file [27]. Building survey instructions and design guidelines ensure that the gbXML offers an appropriate building representation. A series of tools developed (in VB.NET) bridges the data transfer gap between BIM (REVIT) and BEM (EnergyPlus). The tools developed offer editing capabilities for the gbXML file and a novel conversion method from BIM to BEM building representation. Using this process energy performance analysis can be performed for each of the retrofit alternatives. The EplusKPI tool has been developed (in VB.NET) to calculate the KPI values for each of the retrofit alternatives.

Figure 15 shows the input and output files of the EplusKPI Tool. The input files are the gbXML and csv files for each retrofit alternative under exploration. The energy analysis results contained in the EnergyPlus output file (csv) and information stored in the BIM representation of the building (gbXML) are being used to calculate the thermal comfort, indoor air quality, energy consumption, CO2 emissions, Onsite Energy Ratio and Return on Investment KPIs. The output of the tool is an xml file containing the calculated KPI values.

Figure 15 The EplusKPI Tool for calculation of retrofit alternatives KPIs, the input and

output files


2017-07-23

7.2 Checking and filtering out alternatives using benchmarks In Figure 16 the D4E user interface for filtering of the retrofit alternatives can be seen. Each graph displays the calculated (using EnergyPlus and the EplusKPI Tool) values for a specific KPI for all the retrofit alternatives explored, using coloured columns. The x-axis displays the name of the retrofit alternative and the y-axis shows the KPI values together with the relevant unit. A slide-bar to the left of each graph can be used by the user to determine the allowable range of values for each KPI. The slide-bar values are informed by the target benchmark value for each KPI, as set by the user in the Target setting tool. Based on the preferred range of values specified in the slide-bar for each KPI, the retrofit alternatives are filtered out of the ranking and decision making process. This ensures that retrofit alternatives that do not meet the expectations in terms of target benchmark values are not unnecessarily taken into consideration and reduces the computational time.

Figure 16 Checking and filtering out alternatives using benchmarks in the D4E platform


2017-07-23

7.3 Import of KPI weighting Weights for each KPI are used to represent the user preference in terms of criteria prioritisation in the decision making process. The weights for each KPI are set in the D4E platform using the Target Setting tool as explained in chapter 6. Figure 17 presents the export of the tool containing the user-defined weights and target benchmark values for each of the KPIs. The export of the Target Setting tool is an xml file which adheres to a schema developed as part of the Target Setting tool development.

The parent node is the KPItarget node. The KPIs are listed under different wider criteria groups: the economic performance criteria incorporates the Return on Investment KPI under the profitability section; the performance in use criteria includes the Indoor Air Quality and Thermal Comfort KPIs under the indoor environment quality section; the environmental performance criteria links to the Carbon Footprint KPI under the climate change section and to the Non-renewable Primary Energy KPI under the resource depletion section; and finally the impact on neighbourhood performance relates to the Onsite Energy Ratio KPI under the energy section. For each of the KPIs the weight and the target score (A to E) is given. In addition, the target benchmark values are provided as the attribute ‘value’ of the child node ‘Metric’ of each KPI node.

Figure 17 User-defined weighting as exported from the Target Setting Tool of the D4E

platform


2017-07-23

8. RANKING PROCESS OF ALTERNATIVES

8.1 Topsis method The acronym TOPSIS stands for the Technique for Order of Preference by Similarity to Ideal Solution, which is a multi-criteria decision support technique. It was first developed by Hwang and Yoon in 1981 [19] then improved by Yoon in 1987 [20], and Hwang, Lai and Liu in 1993 [21]. TOPSIS is based on the concept that the chosen alternative should have the shortest geometric distance from the positive ideal solution and the longest geometric distance from the negative ideal solution. It is a method of compensatory aggregation that compares a set of alternatives by identifying weights for each criterion, normalising scores for each criterion and calculating the geometric distance between each alternative and the ideal alternative, which is the best score in each criterion. The mathematical background for normalising the decision matrix, identifying the ideal points, calculating the distance of each retrofit alternative from and the relative closeness to the ideal points is given below.

Constructing the normalized decision matrix whose elements are defined as 𝑥"#:

𝑥"# =𝑤#𝑦"#

𝑦"#'(")*

Where:

𝑖 is the index for the retrofit alternative

𝑗 is the index for the KPI

𝑦"# is the KPI value

𝑤# is the weight for the KPI (user defined)

𝑥"# is weighted and normalised

Consequently each attribute 𝑥"#has the same unit length, this is the normalized value of the specific criteria j, for the specific alternative i.

Ideal and negative ideal points:

𝑎∗ is the maximum (minimum) 𝑥"# for each KPI that needs to be maximised (minimised) across all alternatives

𝑎/ is the minimum (maximum) 𝑥"# for each KPI that needs to be maximised (minimised) across all alternatives e.g. of maximised KPI, hours of temperatures within the thermal comfort limit

e.g. of minimised KPI, heating energy demand per floor area

Distance from ideal points:

𝑆"∗ = (𝑥"# − 𝑥#∗)'4#)* is the distance of an alternative from the ideal point across all criteria


2017-07-23

𝑆"/ = (𝑥"# − 𝑥#/)'4#)* is the distance of an alternative from the negative ideal point across

all criteria

Relative closeness to the ideal point:

𝐶"∗ =678

(678967

∗) where the higher the C value the better the alternative

8.2 Ranking process and algorithms A ranking tool has been developed (in VB.NET) following the TOPSIS MCDM methodology to provide the ranking of the retrofit alternatives.

Figure 18 shows the input and output files of the ranking tool. There are two inputs to the ranking tool: the KPI values for all the retrofit alternative options that are within the user-defined range of acceptable values as specified previously (filtered alternatives); and the user-defined weights. Using the TOPSIS methodology explained in section 8.1, out of all the retrofit alternatives explored the best performing options in terms of KPI values are identified. The ideal and negative ideal points are determined as the options that present the maximum KPI values for the KPIs that need to be maximised (such as the RoI KPI) and the minimum KPI values for the KPIs that need to be minimised (such as the CO2 emissions KPI). The remaining retrofit alternative options are ranked in order of closeness to the ideal and negative ideal points using appropriate indexes. The output of the tool is an xml file containing the indexes calculated. The ranking tool output is used by the D4E platform to display the retrofit alternatives under exploration in order of optimality and closeness to the user-defined expectations.

Figure 18 The ranking tool, the input files and the tool output


2017-07-23

The programming language used for the development of the tools is C# [22], ASP.NET [23] and Visual Basic.NET [24] of the .NET framework. Figure 19 shows an example of tool development for the ranking tool. As shown in Figure 19, Microsoft’s integrated development environment (IDE) Visual Studio Community 2015 [25] was used.

Figure 19 Ranking tool algorithm in Visual Studio using VB.NET


2017-07-23

9. DECISION SUPPORT TOOL GUI

9.1 Background on GUI for DST outcome Graphical user interface (GUI) has become an integral technology into decision-maker's and/or user's daily tasks. Decision-makers need to access the information including the outcome of their optimisation work on their PCs and mobile devices to approve or disapprove the final result. As shown in the following figures, the decision support model was proposed by reflecting the factors considered in the development and the detailed description. Various energy retrofit strategies are established. These scenarios apply the basic energy model to create various energy retrofit models (see Figure 20). This figure shows a typical GUI for decision making.

Figure 20 Graphical user interface of the proposed model (part1)

By analysing the energy-saving effects of various energy retrofit models, the potential of achieving the CERT by energy source in each scenario is assessed. Then, by conducting an economic and environmental assessment of these scenarios, the optimal energy retrofit strategy is selected (see Figure 21).


2017-07-23

Figure 21 Graphical user interface of the proposed model (part 2)

In the REFIT energy consumption report on a study conducted by the team al Loughborough University on 20 UK houses which is also based on the monitored gas and electricity data. During which a number of figures were calculated and presented to occupants in the form of REFIT energy consumption reports (Figure 14). The reports replicated the figures presented in the Green Deal advice report but were based on smart meter measurements rather than modelled figures using RdSAP. Calculations required for the REFIT reports were annual gas and electricity consumption and visualisations of energy costs over a year, a week and a day, daily heating period and the annual energy savings for the measures presented in the Green Deal advice reports. The calculation method for each of these items is described below. The rationale for creating the REFIT Energy Consumption Report in this way was to provide energy use information in a simple, accessible format, which could be easily understood by the majority of the population. It is important to note that this study did not set out to explore the most appropriate means of visually communicating energy data; however, homeowners were given opportunity to offer feedback on the design and format of reports.


2017-07-23

Figure 22 REFIT Energy Consumption Report

Design practice has been influenced by the changing landscape of human-centred design research. The user-centred design approach, which began in the 1970s and became widespread by the 1990s, proved to be most useful in the design and development of consumer products [66]. The emerging design practices will change what we design, how we design, and who designs. The impact upon the education of designers will be immense. The patterns of change taking place in the transition from a product perspective to the purpose perspective are described more fully in the following sections. The emerging design practices, on the right, centre around people’s needs or societal needs, and require a different approach in which they need to take longer views and address larger scopes of inquiry. The first reference, for example, is currently working on a design research project to explore the unmet needs and dreams of the growing number of people.


2017-07-23

Figure 23 A snapshot in time of traditional and emerging design practices[66].

Sometimes ‘users’ can play co-creating roles throughout the design process, i.e. become co-designers, but not always. It depends on level of expertise, passion, and creativity of the ‘user’. All people are creative but not all people become designers. Four levels of creativity can be seen in people’s lives: doing, adapting, making and creating (see Figure 24). These four levels vary in terms of the amount of expertise and interest needed. Expertise, interest/passion, effort, and returns grow with each level.

Figure 24 Four levels of creativity [66]

In the example of ENLITEN project [67] data visualisation was carried out in student accommodation. Researchers tested three types of energy data visualisation by showing energy data expressed numerically, using analogue dials and with emotional faces. Energy use figures were compared with a period where no energy visualisation was used. An average energy reduction of 7.7% was found. Participants responded best to the display which used emotional faces. Three forms of energy use visualisation were tested by the ENLITEN project [67] (see Figure 25).


2017-07-23

Figure 25 Three forms of energy use visualisation tested by the ENLITEN project [67]

9.2 Development of prototype for visualisation of design alternatives KPIs

Figure 26 demonstrates the prototype of the graphical user interface for the visualisation of the energy performance results in the form of KPIs to support decision making. The list of the alternatives can be found at the top left of the screen, in the ranked order as calculated by the ranking tool. The user (architect and client) can select up to three retrofit alternatives to display. There are four columns for data display. The first column presents the calculated KPIs for the base-case building, the building as-is before any retrofit actions are taken. In


2017-07-23

each of the remaining three columns the calculated KPIs are presented for one of the selected retrofit alternatives. In Figure 26 the results for the thermal comfort KPI can be seen. Plots of the minimum and maximum temperatures in each of the building zones are provided. The percentile difference from the base-case building is also given to enable easier comparability.

Figure 26 Prototype GUI of the DST KPIs

9.3 Validation of the DST GUI prototype with end users Feedback from professional architects during the project training workshop in Warsaw (Poland) in March 2017 can be summarised in the following points:

• Payback time is usually a preferred metric instead of the Return on Investment KPI for Architects in real life. However, short-term benefits are better explored when using RoI on an annual level.

• The percentile difference between each of the KPIs instead of an overall percentile change would be more beneficial to the Architects when exploring the options with the client.

• Benchmark values for each of the KPIs could be displayed in the GUI and used as a reference of the potential of the actual performance of each of the retrofit alternatives.

• The user interface could be less ‘crowded’ by allowing the user to select the KPIs to visualise instead of presenting all graphs at once.

• The axis scaling should be equal across the different graphs to allow for easier comparability across the different alternatives.

• Overall, the GUI is found useful in terms of data availability to inform decision making processes.


2017-07-23

10. IMPLEMENTATION OF THE RETROFIT AND MAINTENANCE DST IN REAL WORLD CASE STUDY

Once the tools for decision making have all been developed they are tested on a real-world case study which is the GSM building. The D4E energy decision support is meant to be an automated environment where an Architect leads the whole process from generating a BIM file (in REVIT) of an existing building as a base case and undertake a number of activities on the D4E platform to propose an optimised retrofit solution. After producing the BIM file the Architect will define the parameters for generating combinations of retrofit alternatives. The D4E-DST environment will edit the base case file of the building using the Architect defined parameters which combines the selected number of chosen components into all possible mathematical combinations. Then the Architect will retrieve a ZIP file which contains BIM files of edited base case with the new combinations of components. The ZIP file will contain all retrofit alternatives which the Architect will then upload into the ranking process that will calculate the KPIs for each alternative. After that the D4E-DST will proceed with ranking the alternatives and display on a GUI the performance results for the top few alternatives. At this stage, the architect will bring in his expertise to make a final decision using the proposed few solutions that do take into account all the conditions from the user and benchmarks etc.

10.1 GSM building Revit model Figure 27 presents the GSM case study building. The exact specifications for the building have been included in D8.1. The GSM building is a domestic house located in Madrid, Spain. It is a representative example of a detached urban Spanish house in need for retrofitting. The building is occupied by a single-family.

The GSM building was used for demonstration and process evaluation purposes in D5.3. In D5.3 the REVIT model was prepared using relevant guidelines and was transformed in an idf file ready for energy performance analysis using EnergyPlus. The KPIs were calculated for the base-case building (as-is case). In this report, the base-case building is used to create a number of retrofit alternatives using the Options Generator and the components included in the D4E component database. For each of the retrofit alternatives the KPI values are calculated. The weights and target benchmarks are defined by the user using the Target Setting tool user interface provided by the D4E platform. Based on the selection of the benchmarks the user can filter out some of the retrofit alternative options that deviate significantly from the target values according to their preference. The ranking tool is then used to identify the best retrofit alternatives form all possible options. Finally, the results are displayed in the dedicated GUI.


2017-07-23

Figure 27 The GSM case study

10.2 Generation of alternatives The process was tested based on the identification of relevant technologies, starting by the selection of components all the way to the generation of alternatives in gbXML in the options generator. The GSM case study building was used based on the data provided as a model for the options generator platform. The generation of alternative building solutions or “options”, which are analysed and matched to the target values, is supported by a tool, which developed for the D4E platform in the context of the decision support tool. This tool is web-based and strongly interconnected with the building component database. At first, the case study building model is uploaded to the options generator tool. Secondly, different building components such as windows, doors or walls are selected from the component database. The component database has a tool integrated to generate automatically a gbXML-file, which is a development of the XML standard, and draws information from the collaborative database. This gbXML-file is necessary to simulate the building model in EnergyPlus. Thirdly, the options generator creates different building models each one with a different setting of components, which were selected in the component database. The set of building models is automatically provided as a download in a compressed zip-file. Figure 28 shows screenshots of each of these steps.


2017-07-23

Figure 28 Generation of alternatives in the D4E alternatives generator

The implementation of the retrofit and maintenance DST was done with the example of different building components of window, door and wall construction and their (seven) combinations. Resulting in a list of seven gbXML models for further simulation and the calculation of KPIs for retrofitting alternatives compressed in a zip-File.

10.1 Calculating KPIs for retrofit alternatives Figure 29 shows how the gbXML files of retrofit alternatives generated by the Options Generator tool are used to calculate the KPIs that are used for retrofit decision-making. Using the ‘Conversion tool’ developed and presented in D5.3, each gbXML file is converted into an idf file, resulting in seven idf files each containing an EnergyPlus building representation which can be used to proceed with building energy performance analyses. Using EnergyPlus energy performance results are obtained in a csv format. The csv files are then used by the ‘EplusKPI tool’ together with the relevant gbXML files and the base-case building to calculate the KPIs for each retrofit alternative.


2017-07-23

Figure 29 Using the alternative gbXML files to calculate the KPIs

Figure 30 presents the KPI values for one example retrofit alternative as calculated following the process explained above and using the ‘EplusKPI tool’. Similar files are created for each retrofit alternative. There are two types of files containing information regarding the KPI values: an extensive version providing KPI values on different levels of resolution (e.g. annual and monthly values, KPIs broken down into end uses etc.); and a shorter version in which only the annual values for the six KPIs are summarised.

Figure 30 The calculated KPIs in the kpiXML and short kpiXML files


2017-07-23

10.2 Filtering of retrofit alternatives The filtering tool of retrofit alternatives is meant to support narrowing down to the most promising options by eliminating those possibilities that are either too far from satisfying the benchmark data or in conflict with the user requirement as given in the weighting process. However, as we did not develop a large enough data base for creation of an overwhelming set of possible options for retrofit (using the options generator), the project team did not therefore develop or test the suggested filtering tool. It virtual consist of checking whether the values of the KPIs in a possible alternative are within the range or near enough to the values as set in the benchmarks. Similarly, it can also be used to check whether the values of th ekpis are with the preferences of the client if known. The filtering tool also is meant to offer an advantage of performing the decision making process within acceptable time as it reduces the number of possibilities to analyse the processing time drops significantly


2017-07-23

10.3 Ranking alternatives and presentation of results Figure 31 shows the selected (by the user) weights for each of the KPIs under consideration. There are four categories under which the six KPIs are grouped: the economic performance group of KPIs; the performance in use group of KPIs; the environmental performance group of KPIs; and the impact on neighbourhood performance group of KPIs. In this particular case the highest importance was assigned to the ‘Economic performance’ category with a weight of 40 out of 100. Since the only economic performance KPI used is the ‘ROI KPI’, the resulting weight is 40 (100 · =>

*>>= 40). A smaller weight of 20 out of 100 was assigned to the

‘Performance in use’ category. Out of the two KPIs under this category the ‘Thermal Comfort KPI’ was considered most significant with a relative weight of 60 and a weight of 40 for the ‘Indoor Air Quality KPI’, resulting in overall weights of 12 (60 · '>

*>>= 12) and 8 (40 · '>

*>>=

8) out of 100 respectively. The ‘Environmental performance’ category was assigned a weight of 20 out of 100 and relates to the ‘Carbon Footprint KPI’ and the ‘Primary Energy Use KPI’ with weights of 30 and 70 respectively. The resulting overall weights for the two KPIs are 6 (30 · '>

*>>= 6) and 14 (70 · '>

*>>= 14) respectively. Finally, the ‘Impact on Neighbourhood’

category links only to the ‘Onsite Energy Ratio KPI’ with an overall weight of 20 out of 100. Therefore, the most significant KPI in order of importance for this example of weight assignment is the ‘ROI KPI’ (40/100), followed by the ‘Onsite Energy Ratio KPI’ (20/100), the ‘Primary Energy Use KPI’ (14/100), the ‘Thermal Comfort KPI’ (12/100), the ‘Indoor Air Quality KPI’ (8/100) and lastly the ‘Carbon Footprint KPI’ (6/100).

Figure 31 User defined weights for ranking


2017-07-23

Figure 32 using the user defined weights and the KPI values in the kpiXML files and following the TOPSIS methodology the magnitude of closensess to the ideal solutions is calculated for each of the seven retrofit alternatives. The results are included in an xml file.

Figure 32 The magnitude of closeness to the ideal solutions used for ranking of the retrofit

alternatives using TOPSIS

Figure 33 presents the results in the prototype GUI which can be used for decision making. In this case, the three highest ranked alternatives are displayed. The user can use the base-case building results as reference to better understand and assess the resulting values for the alternatives. The importance of the methodology is highlighted here. Without the ranking of the retrofit alternatives it would be extremely difficult for the user to make an informed decision taking into consideration his preferences by just viewing the results. This is because in most KPIs a not so significant improvement from the base-case model is seen. In addition, in most cases, any improvements in some KPI values are counteracted by a decline in other KPI values. Using a ranking process that takes into consideration the user preferences and the availability of the ranking information when exploring the KPI values can be key in enabling a better informed retrofitting decision. In this example of the retrofit decision making tool implementation, the highest ranked alternative related to only replacing the window elements. The second highest ranked alternative related to replacing the wall construction and the third best alternative related to replacing the windows and the doors components. The ‘ROI KPI’ was selected by the user as the most significant KPI. It is evident that this selection had a great influence in the ranking of the alternatives. Out of the three alternatives, the highest ranked alternative presented the most significant ‘ROI KPI’ value (0.24% for the 1st alternative, 0.06% for the second and 0.1% for the third alternative). Without the ranking information, available to highlight the importance of this KPI value, this retrofit alternative might not have been identified as the best option available as improvements in the other KPI values were not as evident. This was the case in the ‘Primary Energy Consumption KPI’ for which the highest ranked alternative presented an improvement in the cooling energy consumption with a decline from 14.8kWh/m2/pa in the base-case model to 10.58 kWh/m2/pa and a simultaneous rise in the heating energy demand from 153.4314.8kWh/m2/pa to 161.5214.8kWh/m2/pa. At the same time, the highest ranked alternative did not present any apparent improvements in the ‘Primary Energy Consumption KPI’ against the other alternatives either. However, this KPI had a much lower weight assigned than the ‘ROI KPI’ (14 to 40) a fact that had a significant impact on the resulting ranking score. It is apparent that the ranking process can be used effectively in conjunction with the detailed KPI values to enable better informed decision making processes for the users.


2017-07-23

Figure 33 Display of KPI results for three best alternatives using the prototype UI


2017-07-23

11. CONCLUSIONS

11.1 Summary of achievements The main aim of this task, which is the development of the decision support tool was achieved in the context of D5.4. It involves: (i) development of prototype tools to automatically generate maintenance/ retrofit advice at different levels of abstraction and for use by different stakeholders, (ii) incorporating ICT visualisation and user-centred design techniques; (iii) creation of strong engagement with the various stakeholders (including industry, consumer and policy) to refine the key features of the retrofit decision support tool and the delivery mechanisms for retrofit or maintenance advice.

A key success factor for further development and application of the solutions developed here is the contribution of maintenance and retrofit stakeholders in the population of both the components database and the benchmarking database. This is strongly linked to the integration of the solutions developed in the construction community on a regional, national and international level.

11.2 Relation to continued developments The results described in D.5.4 are within the core of the D4E project. The decision support system allows stakeholders involved in early design phases of retrofitting and maintenance projects of districts and buildings to early identify relevant technologies and components and generate design alternatives that can be compared based on a set of selected key performance indicators and in relation to existing benchmarking buildings. It is based on information available in the components and energy database developed in WP3 and is envisioned to be integrated into the virtual workspace for holistic energy optimisation for buildings worked upon in WP7.

11.3 Other conclusions and lessons learned Early design phases in the design of maintenance and retrofitting projects of districts and buildings are critical to building and district performance throughout their life-cycle. By empowering all stakeholders to get on the one hand most recent information on technologies and components in a comprehensive and semi-standardised way and on the other hand directly integrate this information along the construction value chain for the generation of measurable alternatives as well as models that can be used for simulation, is able to considerably transform stakeholder integration as we know it today and lead to more efficient and effective buildings and districts.

Being a technology oriented project focused on the development of new solution and prototypes, the D4E project is able to provide these solutions as a key output. What has been learned throughout the project is that it is still a major challenge to master the social perspective and to establish a collaborative tool, enabling all stakeholders to participate in decision making, into the construction and more especially into the maintenance and retrofitting value chain.


2017-07-23

12. ACRONYMS AND TERMS AHP ............................ Analytical hierarchy process

BEM ............................ Building energy model BIM ............................. Building information model

CAD ............................ Computer aided design D .................................. Deliverable

D4E ............................. Design4Energy DST ............................. Decision support tool

EERS ........................... Energy efficient retrofit score EPBD .......................... Energy performance of buildings directive

EPC ............................. Energy performance certificate ERDP .......................... Energy retrofit decision process

ERIDSS ....................... Energy retrofit intelligent decision support system FHR ............................. Fraunhofer

GUI ............................. Graphical user interface ICT .............................. Information and communication technologies

IDSS ............................ Intelligent decision support system IFC .............................. Industry foundation class

KB-EAI ....................... Knowledge-based expert advice on installation KB-HVAC .................. Knowledge-based heating, cooling and air conditioning measures

KB-HWH .................... Knowledge-based hot water heating measures KB-Light ..................... Knowledge-based lighting measures

KB-SES ....................... Knowledge-based stand-alone and energy saving measures KB-Therm ................... Knowledge-based thermal envelope measures module

KMS ............................ Knowledge-based management subsystems KPI .............................. Key performance indicator

LCC ............................. Life-cycle cost LU ............................... Loughborough University

MCDM ........................ Multi criteria decision making ORME ......................... Office rating methodology

PC ................................ Personal computer RoI............................... Return on investment

TFN ............................. Triangular fuzzy numbers VB ............................... Visual Basic


2017-07-23

WDM .......................... Weighted product model

WP............................... Work package WSM ........................... Weighted sum model

XML ............................ Extensible markup language


2017-07-23

13. REFERENCES [1] Agrawal, A., 2015. Qualitative decision methods for multi-attribute decision

making. arXiv preprint arXiv:1508.00879. [2] Balaras, C.A., et al., 2004. Decision support software for sustainable building

refurbishment. Transactions-American Society of heating Refrigeration and Air Conditioning Engineers, 110(1), pp.592-601.

[3] European Commission. EC Directorate general for energy and transport. Intelligence energy – Europe Programme. Multiplying success in buildings: 21 innovative projects for an energy-intelligent Europe. Energy performance assessment for existing non-residential buildings—EPA-NR. Brussels, 2006a.

[4] European Commission. EC Directorate General for energy and transport. Intelligence Energy–Europe Programme. Multiplying success in buildings: 21 innovative projects for an energy-intelligent Europe. Collecting data from energy certification to monitor performance indicators for new and Existing buildings—DATAMINE. Brussels, 2006b.

[5] Eastman, C.M., et al., 2011. BIM handbook: A guide to building information modelling for owners, managers, designers, engineers and contractors. John Wiley & Sons.

[6] Dubé, L., et al., 2005. The impact of structuring characteristics on the launching of virtual communities of practice. Journal of Organizational Change Management, 18(2), pp.145-166.

[7] Gero, J.S., et al., 1983. Energy in context: a multicriteria model for building design. Building and Environment, 18(3), pp.99-107.

[8] Jaggs, M. and Palmer, J., 2000. Energy performance indoor environmental quality retrofit—a European diagnosis and decision making method for building refurbishment. Energy and buildings, 31(2), pp.97-101.

[9] Succar, B., 2009. Building information modelling framework: A research and delivery foundation for industry stakeholders. Automation in construction, 18(3), pp.357-375.

[10] Kumbaroğlu, G. & Madlener, R., 2012. Evaluation of economically optimal retrofit investment options for energy savings in build.. Energy & Buildings, 49, pp.327-334.

[11] Costa, A., et al., 2012. Key factors methodology — A novel support to the decision making process of the building energy manager in defining optimal operation strategies. Energy and Buildings, 49, pp.158-163

[12] Diakaki, C., et al., 2010. A multi-objective decision model for the improvement of energy efficiency in buildings. Energy, 35(12), pp.5483-5496.

[13] Kaklauskas, A., et al., 2005. Multivariant design and multiple criteria analysis of building refurbishments. Energy and Buildings, 37(4), pp.361-372.

[14] Alanne, K., 2004. Selection of renovation actions using multi-criteria “knapsack” model. Automation in Construction, 13(3), pp.377-391.

[15] Diakaki, C., et al., 2008. Towards a multi-objective optimization approach for improving energy efficiency in buildgs.. Energy and Buildings, 40(9), pp.1747-1754.

[16] Asadi E, et al. Building and Environment 56 (2012) 370-378 [17] Ibn-Mohammed T, et al. / Building and Environment 72 (2014) 82e10 [18] Dimitriou V, Firth SK, Hassan TM, Fouchal F, Oraiopoulos A, and Malo P, 2017.

Deliverable 5.3 - Data analytics based on building energy performance modelling. [19] Hwang, C.L.; Yoon, K. (1981). Multiple Attribute Decision Making: Methods and

Applications. New York: Springer-Verlag.


2017-07-23

[20] Yoon, K. (1987). A reconciliation among discrete compromise situations. Journal of Operational Research Society. 38. pp. 277–286.

[21] Hwang, C.L.; Lai, Y.J.; Liu, T.Y. (1993). "A new approach for multiple objective decision making". Computers and Operational Research. 20: 889–899.

[22] Microsoft, 2017. [Online]<Available at: https://msdn.microsoft.com/en-gb/library/67ef8sbd.aspx >[Date accessed: 28/03/2017]

[23] Microsoft, 2017. [Online]<Available at: https://www.asp.net/ >[Date accessed: 11/01/2017]

[24] Microsoft, 2017. [Online]<Available at: https://msdn.microsoft.com/en-us/library/aa903378(v=vs.71).aspx >[Date accessed: 11/01/2017]

[25] Microsoft, 2017. [Online] <Available at: https://www.visualstudio.com/vs/community >[Date accessed: 11/01/2017]

[26] Hyvärinen, J., et al., 2016. Deliverable 2.2b - Definition of usage scenarios with economic and environmental criteria.

[27] Dimitriou V, Firth SK, Hassan TM, Fouchal F, Oraiopoulos A, and Malo P, 2017. Deliverable 5.3 - Data analytics based on building energy performance modelling.

[28] Juan, Y.K., et al., 2009. GA-based decision support system for housing condition assessment and refurbishment strategies. Automation in Constr., 18(4), pp.394-401.

[29] Zeleny, Milan. Multiple criteria decision making. Ed. J. L. Cochrane. Vol. 25. NY McGraw-Hill, 1982.

[30] Duah, D. and Syal, M.. Intelligent decision support system for home energy retrofit adoption. International Journal of Sustainable Built Environment. Vol. 5. Elsevier B.V., 2016. pp.620-634.

[31] Kolokotsa, D. et al.. Decision support methodologies on the energy efficiency and energy management in buildings. Vol. 3. Earthscan, 2009. pp.121-146.

[32] Florides, G. A. et al.. Measures used to lower building energy consumption and their cost effectiveness. Applied Energy. Vol. 73, no 3-4. 2002. pp.299-328.

[33] Zurigat, Y. H.. Energy efficient building strategies for school buildings in Oman. International Journal of Energy Research. Vol. 27, no 3. 2003. pp.241-253.

[34] Becker, R. et al.. Improving energy performance of school buildings while ensuring indoor air quality ventilation. Building and Environment. Vol. 42, no 9. 2007. pp.3261-3276.

[35] Tavares, P. F. A. F. and Martins, A. M. O. G.. Energy efficient building design using sensitivity analysis – A case study. Energy and Buildings. Vol. 39, no 1. 2007. pp.23-31.

[36] Fouchal, F. et al.. Decision support to enable energy efficient building design for optimized retrofit and maintenance. Proceedings of the 32nd CIB W78 Conference, Eindhoven, The Netherlands, 27-29 October 2015, pp.205-214.

[37] Saaty, Thomas L.. Fundamentals of decision making and proproty theory: with the analytic hierarchy process. Rws Publications, 1994.

[38] Roulet, C. A. et al.. ORME: A multi-criteria rating methodology for buildings. Building and Environment. Vol. 37, no 6. 2002. pp.579-586.

[39] Pasanisi, A. and Ojalvo, J.. A multi-criteria decision tool to improve the energy efficiency of residential buildings. Foundations of Computing and Decision Sciences. Vol. 33, no 1. 2008. pp.71-82.

[40] Kolokotsa, D. et al.. Implementation of an integrated indoor environment and energy management system. Energy and Buildings. Vol. 37, no 1. 2005. pp.93-99.

[41] Kolokotsa, D.. Comparison of the performance of fuzzy controllers for the management of the indoor environment. Building and Environment. Vol. 38, no 12. 2003. pp.1439-1450.


2017-07-23

[42] Kolokotsa, D.. Artificial intelligence in buildings: A review on the application of fuzzy logic. Advances in Building Energy Research. Vol. 1. 2007. pp.29-54.

[43] Kalogirou, S. A.. Artificial neural networks in energy applications in buildings. International Journal of Low Carbon Technologies. Vol. 1, no 3. 2006. pp.201-216.

[44] Ma, Z. et al.. A supervisory control strategy for building cooling water systems for practical and real time applications. Energy Conversion and Management. Vol. 49, no 8. 2008. pp.2324-2336.

[45] Grillo, Maria Cristina et al.. Industrial estate retrofitting: Selection of sustainable strategies using MCA. Central Europe towards Sustainable Building. 2013. Decision-support tools and assessment methods.

[46] OpenStudio. [Online]<Available at: http://openstudio.nrel.gov/ >[Date accessed: 06/04/2017]

[47] About NREL’s OpenStudio Application Suite and Development Platform. [Online]<Available at: https://www.youtube.com/watch?v=ovLt4-q_UEg >[Date accessed: 06/04/2017]

[48] eXtensible Building Information Modelling (xBIM) Toolkit. [Online]<Available at: https://github.com/xBimTeam >[Date accessed: 06/04/2017]

[49] Open source BIM Server. [Online]<Available at: http://bimserver.org/ >[Date accessed: 06/04/2017]

[50] Standards from BuildingSMART. [Online]<Available at: http://www.buildingsmart.org/standards/ >[Date accessed: 06/04/2017]

[51] IFD Library & IFD Library Group for BuildingSMART. [Online]<Available at: http://www.ifd-library.org/ >[Date accessed: 06/04/2017]

[52] Fleming, Katherine et al.. The Building Component Library: An Online Repository to Facilitate Building Energy Model Creation. ACEEE Summer Study on EE in Buildings. Pacific Grove, California, August 12-17, 2012.

[53] Building Component Library. [Online]<Available at: https://bcl.nrel.gov/about >[Date accessed: 06/04/2017]

[54] European project ISES. Deliverable D4.1. [Online]<Download at: http://ises.eu-project.info/documents/ISES-D4.1.pdf >[Date accessed: 06/04/2017]

[55] European project ISES. Deliverable D4.3. [Online]<Download at: http://ises.eu-project.info/documents/ISES-D4.3.pdf >[Date accessed: 06/04/2017]

[56] MagiCAD plugins [Online]<Available at: http://www.magicad.com/en/software/plugins-main/ >[Date accessed: 06/04/2017]

[57] ArchiCAD Object Download Sites [Online]<Available at: http://www.graphisoft.com/downloads/parametric_objects.html >[Date accessed: 06/04/2017]

[58] OpenEI: Energy Information. Data and other resource. [Online]<Available at: http://en.openei.org/wiki/ >[Date accessed: 06/04/2017]

[59] European construction information system. euroBAU.com. [Online]<Available at: http://www.eurobau.com/<[ Date accessed: 06/04/2017]

[60] Material data collection for the energetic old building refurbishment. [Online]<Available at: http://www.masea-ensan.com/ >[Date accessed: 06/04/2017]

[61] Material data for the energetic old building refurbishment. [Online]<Available at: http://www.bine.info/publikationen/publikation/materialdaten-fuer-die-energetische-altbausanierung/materialdatenbank/ >[Date accessed: 06/04/2017]

[62] National Residential Efficiency Measures Database by NREL. [Online]<Available at: http://www.nrel.gov/ap/retrofits/ >[Date accessed: 06/04/2017]

[63] Climate-Smart Planning Platform. [Online]<Available at: https://www.climatesmartplanning.org/ >[Date accessed: 06/04/2017]

[64] The largest platform for building information and construction projects worldwide. [Online]<Available at: http://www.emporis.de/ >[Date accessed: 06/04/2017]


2017-07-23

[65] Juan, Y.-K.. Kim, J.H Roper, K. Lacouture, D.C. (2009) GA-based decision support system for housing condition assessment and refurbishment strategies. Automation in Construction, 18, pp. 394–401

[66] Sanders, E. B.-N. & Stappers, P. J. (2008) Co-creation and the new landscapes of design. International Journal of CoCreation in Design and the Arts, Volume 4, 2008 – Issue 1, pp 5-18.

[67] Chiang, T., Mevlevioglu, G., Natarajan, S., Padget, J. and Walker, I., 2014. Inducing [sub]conscious energy behaviour through visually displayed energy information: A case study in university accommodation. Energy and Buildings, 70, pp. 507-515.

[68] Steemers, K., 2003. Energy and the city: Density, buildings and transport. Energy and Buildings. 35 (1), pp. 3–14.

[69] Ratti, C., Baker, N., and Steemers, K., 2005. Energy consumption and urban texture. Energy and Buildings. 37 (7), pp. 762–776.

[70] Jones, P., Patterson, J., and Lannon, S., 2007. Modelling the built environment at an urban scale-Energy and health impacts in relation to housing. Landscape and Urban Planning. 83 (1), pp. 39–49.

[71] Geidl, M., Koeppel, G., Favre-Perrod, P., Klöckl, B., Andersson, G., and Fröhlich, K., 2007. The Energy Hub – A Powerful Concept for Future Energy Systems. In: Third Annual Carnegie Mellon Conference on the Electricity Industry.

[72] Fonseca, J.A. and Schlueter, A., 2015. Integrated model for characterization of spatiotemporal building energy consumption patterns in neighbourhoods and city districts. Applied Energy. 142, pp. 247–265

[73] Fonseca, J.A., Nguyen, T.A., Schlueter, A., and Marechal, F., 2016. City Energy Analyst (CEA): Integrated framework for analysis and optimization of building energy systems in neighbourhoods and city districts. Energy and Buildings. 113, pp. 202–226.

[74] Robinson, D., Haldi, F., Kämpf, J., Leroux, P., Perez, D., Rasheed, A., and Wilke, U., 2009. CitySIM: Comprehensive micro simulation for resource flows for sustainable urban planning. In: Eleventh International IBPSA Conference. Glasgow, Scotland. pp. 1083–1090.

[75] Walter, E. and Kampf, J., 2015. A verification of CitySim results using the BESTEST and monitored consumption values. In: Proceedings of the 2nd Building Simulation Applications conference. Bozen-Bolzano.

[76] Dorer, V., Allegrini, J., Orehounig, K., Moonen, P., Upadhyay, G., Kämpf, J., and Carmeliet, J., 2013. Modelling the urban microclimate and its impact on the energy demand of buildings and building clusters. In: Proceedings of BS 2013. France. pp. 3483–3489.

[77] Thomas, D., Miller, C., Kämpf, J., and Schlüter, A., 2014. Multiscale Co-Simulation of Energyplus and Citysim Models derived from a Building Information Model. In: Proceedings of BauSIM 2014: 5th Germ an-Austrian Conference of IBPSA. Aaechen. pp. 469–476.

[78] Miller, C., Thomas, D., Kämpf, J., and Schlüter, A., 2015. Long Wave Radiation Exchange for Urban Scale Modelling within a Co-simulation Environment. In: Proceedings of International Conference CISBAT 2015 “Future Buildings and Districts Sustainability from Nano to Urban Scale. Lausanne. pp. 871–876.

[79] Nouvel, R., Brassel, K.-H., Bruse, M., Duminil, E., Coors, V., Eicker, U., and Robinson, D., 2015. SimStadt, a new worflow-driven urban energy simulation platform for CityGML city models. In: International Conference CISBAT 2015 “Future Buildings and Districts – Sustainability from Nano to Urban Scale. Lausanne.


2017-07-23

[80] Hong, T., Chen, Y., Lee, S.H., and Piette, M.A., 2016. CityBES : A Web-based Platform to Support City- Scale Building Energy Efficiency. In: Urban Computing. San Francisco.

[81] Gröger, G., Kolbe, T.H., Nagel, C., and Häfele, K.-H., 2012. OGC City Geography Markup Language (CityGML) Encoding Standard.

[82] Kolbe, T.H., 2009. Representing and Exchanging 3D City Models with CityGML. In: J. Lee and S. Zlatanova, eds. 3D Geo-Information Sciences. Berlin: 3D Geo-Information Sciences. pp. 15–32.

[83] Nouvel, R., Kaden, R., Bahu, J.-M., Kaempf, J., Cipriano, P., Lauster, M., Benner, J., Munoz, E., Tournaire, O., and Casper, E., 2015. Genesis of the CityGML Energy ADE. In: International Conference Future Buildings & Districts Sustainability from Nano to Urban Scale : CISBAT2015. Lausanne.

report 5.4 decision support tool...design4energy!d5.4 decision support tool development page 5 of 64...

Documents