business intelligence model (bim) updatejenhork/presentations/bim_torontoselab_july20… ·...
TRANSCRIPT
The Business Intelligence Model Representation, Reasoning, and Application
University of Toronto, SE Lab, July 3, 2012 JENNIFER HORKOFF1, DANIELE BARONE1, LEI JIANG1, ERIC YU2, DANIEL AMYOT3,
ALEX BORGIDA4, JOHN MYLOPOULOS1
1Department of Computer Science, University of Toronto, Canada jenhork, barone, leijiang, [email protected]
2Faculty of Information, University of Toronto, Canada [email protected]
3EECS, University of Ottawa, Canada [email protected]
4Department of Computer Science, Rutgers University, USA [email protected]
Business Intelligence Model 1
Outline Business Intelligence Model Aims
Previous Work (“Stage 1”)
BIM Concepts and Reasoning Consolidation (“Stage 2”) Hybrid Reasoning
BIM Language Semantics and Reasoning (“Stage 3”) Metamodel
Definition and metaproperties
Additional Reasoning Capabilities
BIM in Action: A Hospital Case Study
Conclusions
Business Intelligence Model 2
Business Intelligence (from Wikipedia) Business intelligence (BI) is the ability for an
organization to take all its capabilities and convert them into knowledge, ultimately, getting the right information to the right people, at the right time, via the right channel.
This produces large amounts of information which can lead to the development of new opportunities for the organization.
When these opportunities have been identified and a strategy has been effectively implemented, they can provide an organization with a competitive advantage in the market, and stability in the long run (within its industry).
3
Interfacing.com
Pentaho.com
HighJump.com
Business Intelligence Model
Business Intelligence Model Aims BI Systems are widely used, but
Systems are still very technical and data-oriented
Hard to understand what the data means
Hard to design queries or make new reports without technical knowledge or a knowledge of the underlying data structure
Gap between business and IT-supplied data
Business people would rather reason using their own terms: Strategic objectives, business models and strategies, business
processes, markets, trends and risks
Raise the level of abstraction of BI systems using a modeling language Uses concepts more familiar to business
Business Intelligence Model 4
Business Intelligence Network BIM is part of the Business Intelligence Network*, a
Canadian project for the definition of the next generation of Business Intelligence Technologies.
*http://bin.cs.toronto.edu
Business Intelligence Model 5
Business Intelligence Model (BIM) Development
May existing languages and techniques for capturing business strategy Strategy Maps and Balanced Scorecards (Kaplan & Norton)
Business Motivation Model (OMG)
Dynamic SWOT (Strength, Weakness, Opportunity, Threat) Analysis (Dealtry)
Goal Models
These techniques offer many useful concepts, but often not clearly defined visions, objectives, goals, means, strategies, plans, metrics, indicators,
measures, strengths, weaknesses, threats, vulnerabilities, opportunities, etc..
BIM aims to select a consolidated set of core concepts
Business Intelligence Model 6
BIM Development: “Stage 1”
When I joined the project September 2011
BIM Tech report, PoEM’10 Concepts, background, more detail
ER’11: Jiang et al. BIM concepts
Application of existing analysis procedures (goal modeling, decision analysis) through mapping to BIM
ER’11, PoEM’11 Barone et al. Composite indicators
Reasoning with indicators: unit conversion, normalization, performance levels
Business Intelligence Model 7
Barone et al., “Enterprise Modeling for Business Intelligence,” PoEM’10 Jiang et al., “Strategic Models for Business Intelligence: Reasoning About Opportunities and Threats,” ER'11 Barone et al., “Composite Indicators for Business Intelligence,” ER'11 Barone et al., “Reasoning with Key Performance Indicators,” PoEM’11
BIM Concepts and Reasoning Consolidation: “Stage 2”
One consistent description of BIM concepts Merging of concept descriptions from existing papers
Consistent “picture” or narrative of BIM reasoning Introducing hybrid reasoning
Business Intelligence Model 8
Horkoff et al. “Strategic Business Modeling: Representation and Reasoning”, SoSym (to appear)
Consolidated BIM Concepts Goal: an objective of a business
Can be AND/OR refined
Process: achieves goals
Domain Assumption: properties required for goal satisfaction
Situation: internal or external factors influencing fulfillment of goals Could be SWOT for a particular goal
Influence: situations/goals influence situations/goals Can be logical (implication) or probabilistic (P(A|B))
Indicator: performance measure, quantifies aspects of strategic activities (KPI)
Business Intelligence Model 9
Simple BIM Example
Business Intelligence Model 10
Legend
Goal
Indicator
Process
Situation
To open sales
channels
To increase
sales
Increased
competition
High
demand
To increase
sales volume
To maintain
gross margin
-
To offer
promotions
+
To reduce
costs
-
--
Total sales
Sales
volume
Gross
margin
Economic
SlowdownLow cost
financing
-
++
AND
OR
Evaluates
Domain
Assumption
AND
Refinement
OR
+
Influence
Achieves
Less Simple Example
Business Intelligence Model 11
To open sales
channels
To increase
sales
Increased
competition
To acquire
technology through
acquisition
To develop new
technology in-house
To reduce
risks
To reduce
patent
infringement
lawsuit risk
To reduce
external
dependence
To establish
strategic
partnership
-
To reduce
financial risk
+
+
To invest in new
technologies
-
To maintain
competitive
advantage
+
+
Acquire a
competitor
Develop a
technology
+
++Sufficient
funds
Strong R & D
capability
High
demand
To maintain
revenue growth
To increase
sales volume
To maintain
gross margin
-
To offer
promotions
-+
To reduce
costs
-
-
-
Total sales
Sales
volume
Gross
margin
-
Economic
Slowdown
Healthy
balance sheet High R&D
expenditure Low cost
financing
+++
+
+
-
++
AND
ANDAND
AND
OR
OR
OR
BIM Reasoning
Reasoning with BIM allows an organization to answer strategic or monitoring questions. For example, BestTech may want to pose the following questions: Should we develop technology in-house or acquire technology
through acquisition? Which option is better for maintaining revenue growth and reducing risks?
Is it possible to maintain revenue growth while reducing risks? What strategies can achieve these goals?
Reasoning Technique Required Information
Goal Model Reasoning Initial Reasoning Values
Probabilistic Decision Analysis Conditional Probability Tables, Utility Functions
Reasoning with Indicators Atomic Indicator Values, Business Formulae,
Unit conversion factors
Hybrid Reasoning
(Reasoning with Incomplete
Indicators)
Atomic Indicator Values, (Optional) Business
Formulae, (Optional) Unit conversion factors,
(Optional) Initial Reasoning Values
Business Intelligence Model 12
Reasoning Overview
Business Intelligence Model 13
To open sales
channels
To increase
sales
Increased
competition
To acquire
technology through
acquisition
To develop new
technology in-house
To reduce
risks
To reduce
patent
infringement
lawsuit risk
To reduce
external
dependence
To establish
strategic
partnership
-
To reduce
financial risk
+
+
To invest in new
technologies
-
To maintain
competitive
advantage
+
+
Acquire a
competitor
Develop a
technology
+
++
Sufficient
funds
Strong R & D
capability
High
demand
To maintain
revenue growth
To increase
sales volume
To maintain
gross margin
-
To offer
promotions
-
+
To reduce
costs
-
-
-
Total sales
Gross
margin
-
Economic
Slowdown
Healthy
balance sheet High R&D
expenditure Low cost
financing
+++
+
+
-
++
AND
ANDAND
AND
OR
OR
OR
Sales
volume (i1)
Number of
Sales
Channels (i2)
Number of
Promotions
(i3)
Number of
Competitors
(i4)
?
Re
aso
nin
g w
ith
In
dic
ato
rs
Exis
tin
g R
ea
so
nin
g T
ech
niq
ue
s
Hybrid Reasoning
To open sales
channels
To increase
sales
Increased
competition
To acquire
technology through
acquisition
To develop new
technology in-house
To reduce
risks
To reduce
patent
infringement
lawsuit risk
To reduce
external
dependence
To establish
strategic
partnership
-
To reduce
financial risk
+
+
To invest in new
technologies
-
To maintain
competitive
advantage
+
+
Acquire a
competitor
Develop a
technology
+
++Sufficient
funds
Strong R & D
capability
High
demand
To maintain
revenue growth
To increase
sales volume
To maintain
gross margin
-
To offer
promotions
-+
To reduce
costs
-
-
-
Total sales
Sales
volume
Gross
margin
-
Economic
Slowdown
Healthy
balance sheet High R&D
expenditure Low cost
financing
+++
+
+
-
++
AND
ANDAND
AND
OR
OR
OR
Evaluation of Specific Strategies Should we develop technology in-house or acquire
technology through acquisition?
Business Intelligence Model 14
To open sales
channels
To increase
sales
Increased
competition
To acquire
technology through
acquisition
To develop new
technology in-house
To reduce
risks
To reduce
patent
infringement
lawsuit risk
To reduce
external
dependence
To establish
strategic
partnership
-
To reduce
financial risk
+
+
To invest in new
technologies
-
To maintain
competitive
advantage
+
+
Acquire a
competitor
Develop a
technology
+
++
Sufficient
funds
Strong R & D
capability
High
demand
To maintain
revenue growth
To increase
sales volume
To maintain
gross margin
-
To offer
promotions
-
+
To reduce
costs
-
-
-
Total sales
Sales
volume
Gross
margin
-
Economic
Slowdown
Healthy
balance sheet High R&D
expenditure Low cost
financing
+++
+
+
-
++
AND
ANDAND
AND
OR
OR
OR
PSPD
FD
PS FSPD PD PS
PS
FDPS
PS
Fully
Satisfied/
Denied
Partially
Satisfied/
Denied
FS/FD
PS/PD
Goal Model Reasoning (Giorgini et al.), mapped to BIM
Discovery of Alternative Strategies Is it possible to maintain revenue growth while reducing
risks? What strategies can achieve these goals?
Business Intelligence Model 15
To open sales
channels
To increase
sales
Increased
competition
To acquire
technology through
acquisition
To develop new
technology in-house
To reduce
risks
To reduce
patent
infringement
lawsuit risk
To reduce
external
dependence
To establish
strategic
partnership
-
To reduce
financial risk
+
+
To invest in new
technologies
-
To maintain
competitive
advantage
+
+
Acquire a
competitor
Develop a
technology
+
++
Sufficient
funds
Strong R & D
capability
High
demand
To maintain
revenue growth
To increase
sales volume
To maintain
gross margin
-
To offer
promotions
-
+
To reduce
costs
-
-
-
Total sales
Sales
volume
Gross
margin
-
Economic
Slowdown
Healthy
balance sheet High R&D
expenditure Low cost
financing
+++
+
+
-
++
AND
ANDAND
AND
OR
OR
OR
FSPS
FS
FS
FSFs
FS
Fs
FSFS PS
PD
FD
FD
FDFD
PS
PD
Fully
Satisfied/
Denied
Partially
Satisfied/
Denied
FS/FD
PS/PD
FS
PD
PS
PD
FDFD
PDPD
PD
PD PD
Goal Model Reasoning (Giorgini et al.), mapped to BIM
Probabilistic Evaluation of Strategies
Should we develop technology in-house or acquire technology through acquisition?
Business Intelligence Model 16
Influence diagrams (Howard & Matheson), mapped to BIM
Reasoning with Indicators
Business Intelligence Model 17
To increase
sales volume
Sales
volume
Increase
Sales
Evaluates Measures
Performance
Region
Target
value
Current
value
Threshold
value
Worst
value
Performance level = 0.5
“partially performant”
Satisfaction level = 0.5
“partially satisfied”
Indicator Reasoning with Varying Levels of Information
Reasoning Type Unit Conversion Required Information
Indicator Reasoning using Unit Conversion
Unit conversion factors
Atomic Indicator Values, Business Formulae, Unit
conversion factors
Indicator Reasoning using Performance
Levels
Unit Normalization (Performance
Levels)
Atomic Indicator Values, Business
Formulae
Indicator reasoning without Business
Formula
Unit Normalization (Performance
Levels)
Atomic Indicator Values
Hybrid Reasoning (with Incomplete
Indicators)
Qualitative Normalization
Atomic Indicator Values, (Optional)
Business Formulae, Unit conversion factors, Initial
Reasoning Values Business Intelligence Model 18
Indicator Reasoning using Business Formulae and Unit Conversion
Business Intelligence Model 19
To open new
sales channels
(g2)
To increase
sales volume
(g1)
To offer
promotions (g3)
Sales
volume
(i1)
OR
Number of
Sales
Channels (i2)
Number of
Promotions
(i3)
Increased
Competition
(s1)
-
Number of
Competitors
(i4)
cv(i3) = 5 promotions
cf(i3,i1) = 7cv(i2) = 10 sales channels
cf(i2,i1) = 20
cv(i4)= 2 competitors
cf(i4,i2) = 2
cv = 155 thousand $ in
sales
(𝑐𝑣(𝑖2) − 𝑐𝑣(𝑖4)𝑐𝑓(𝑖4, 𝑖2)) 𝑐𝑓 𝑖2, 𝑖1
+ 𝑐𝑣(𝑖3) 𝑐𝑓 𝑖3, 𝑖1
= 20 𝑐𝑣(𝑖2 − 2𝑐𝑣(𝑖4)) + 7𝑐𝑣(𝑖3)
Indicator Reasoning using Business Formulae and Performance Levels
Business Intelligence Model 20
To open sales
channels (g2)
To increase
sales volume
(g1)
To offer
promotions (g3)
Sales
volume
(i1)
OR
Number of
Sales
Channels (i2)
Number of
Promotions
(i3)
Increased
Competition
(s1)
-
Number of
Competitors
(i4)
pl = -0.1pl = 0.8
pl = 0.2
pl = 0.5
To increase
sales volume
Sales
volume
Increase
Sales
Evaluates Measures
Performance
Region
Target
value
Current
value
Threshold
value
Worst
value
Performance level = 0.5
“partially performant”
Satisfaction level = 0.5
“partially satisfied”
Indicator Reasoning without Business Formulae
Business Intelligence Model 21
To open sales
channels (g2)
To increase
sales volume
(g1)
To offer
promotions (g3)
Sales
volume
(i1)
OR
Number of
Sales
Channels (i2)
Number of
Promotions
(i3)
Increased
Competition
(s1)-
Number of
Competitors
(i4)(per+, per-)
(0.2, 0.0)
-
(per+, per-)
(0.25, 0.0)(per+, per-)
(0.0, 0.1)
(per+, per-)
(0.25, 0.1)
+
-+
+ -
+ -Conflict
Hybrid Reasoning (Reasoning with Incomplete Indicators)
To open sales
channels
To increase
sales
Increased
competition
To acquire
technology through
acquisition
To develop new
technology in-house
To reduce
risks
To reduce
patent
infringement
lawsuit risk
To reduce
external
dependence
To establish
strategic
partnership
-
To reduce
financial risk
+
+
To invest in new
technologies
-
To maintain
competitive
advantage
+
+
Acquire a
competitor
Develop a
technology
+
++Sufficient
funds
Strong R & D
capability
High
demand
To maintain
revenue growth
To increase
sales volume
To maintain
gross margin-
To offer
promotions
-
+
To reduce
costs
-
-
-
Total sales
Sales
volume
Gross
margin-
Economic
Slowdown
Healthy
balance sheet High R&D
expenditure Low cost
financing
+++
+
+
-
++
AND
ANDAND
AND
OR
OR
OR
PSPD
FD
PSPD PD PS
PS
FDPS
PS
(0.4, 0.0)
-
(per+, per-)
(1.0, 0.0)
+
(0.0, 0.25)
(1.0, 0.25)
(0.0, 0.1)
(0.25, 0.0)
(0.2, 0.0)
Number of
Competitors
(per+, per-)
(0.2, 0.0)
-+Number of
Sales
Channels
-
(per+, per-)
(0.8, 0.0)
+
Number of
Promotions
(per+, per-)
(0.0, 0.1)
+ -
(0.4, 0.25)
(0.25, 0.1)
(per+, per-)
(0.5, 0.0)
+ -
PD
Business Intelligence Model 22
BIM Language Semantics and Reasoning (“Stage 3”)
Review concepts introduced in existing business and goal model languages BMM, SWOT, BSc, GMs, SM
Review all concepts previously introduced as part of BIM
Select set of “core” BIM concepts and relationships
Determine how concepts and relationships interact What is allowed, what is not?
Iterative process of language (re)design
Define concepts formally using description logic
Business Intelligence Model 23
Horkoff et a. “Making data meaningful: The Business Intelligence Model and its Formal Semantics in Description Logics”, ODBASE (to appear)
BIM Language Semantics and Reasoning: Metamodel
Consolidated language “metamodel”/upper-level ontology
Business Intelligence Model 24
Business Schema
+name : string
+evidence : Set of EvidenceValue
Thing
+pursued : Set of PursuitValue
Goal
Situation
+target : float
+threshold : float
+currentValue : float
Indicator Task
1*
Relationship1
*
+strength [0..1] : StrengthLabel
+pursue [0..1] : PursuitLabel
Influence
+and : Boolean
Refines Measures
Evaluates
+internal : Boolean
Organizational Situation
+Strong Evidence For = SF
+Weak Evidnece For = WF
+Weak Evidence Against = WA
+Strong Evidence Against = SA
«enumeration»
EvidenceValue
+Pursued = Pur
+Not Pursued = NotPur
«enumeration»
PursuitValue
Entity
+Strong Positive = ++
+Weak Positive = +
+Weak Negative = -
+Strong Negative = --
«enumeration»
StrengthLabel
+Pursue = P
+Not Pursue = !P
«enumeration»
PursueLabel
-source
1 *
1-destination
*
BIM Language Semantics and Reasoning: Definition and Metaproperties
Formal definition of language concepts and relationships Using description logic, e.g.,
Class: Goal SubClassOf: Situation
Property: influences Domain: Situation Range: Situation InverseOf: infBy
More specialized concepts representing using metaproperties duration (long-term/short-term), likelihood of fulfillment
(high/low), nature of definition (formal/informal), scope (broad/narrow), number of instances (many/few), and perspective from BSC (financial/ customer/ internal/ learning and growth)
E.g., Vision is a “goal with a long duration, broad scope, low chance of fulfillment, informal definition, and few instances”
Business Intelligence Model 25
OWL Protégé Implementation
Business Intelligence Model 26
BIM Language Semantics and Reasoning: Additional Reasoning Capabilities
Description of BIM in DL Is easily extensible
Allows publishing of generic BIM models as ontologies on the semantic web
DL allows reasoning capabilities beyond the application of existing approaches in previous work: Can support “What if?” type reasoning from GM, but now
inherent to the language, no mapping
Tested using an OWL encoding in Protégé
Introduced reasoning with pursuit
Detecting inconsistencies in BIM Schemas
Automatically classify defined concepts relative to existing concepts, organizing the model
Allowing more detailed conceptual modeling of entities, tasks, etc.
Business Intelligence Model 27
Business Intelligence Modeling in Action: A Hospital Case Study
Daniele Barone*, Thodoros Topaloglou**, and John Mylopoulos*
*Computer Science Department, University of Toronto, Canada barone, [email protected]
**Rouge Valley Health System, Toronto, Canada [email protected]
Selected slides from…
Abstract We present results of using the Business Intelligence
Model (BIM) in the definition of requirements for a Business Intelligence (BI) Solution currently undertaking at the Rouge Valley Health System (RVHS)
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 29
Business Intelligence
Solution
Business Intelligence
Solution
BIM BIM
RVHS RVHS
A vendor BI
platform
A vendor BI
platform
Case Study Questions and Method Questions:
What is the value of BIM in a BI implementation?
Is the initial BIM language sufficient to support the business modeling needs of the case study?
Who are the users of BIM?
Is there a development methodology that matches with BIM?
How does BIM map to data?
Method: in situ case study in a healthcare organization during BI
implementation.
researchers worked side-by-side with a BI development team.
shadow the implementation effort and generate models that capture the requirements and design choices of the implementation.
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 30
Rouge Valley Health System RVHS is a two site hospital with 479 beds in the east greater Toronto
area
Key facts 2700 employees
Over 500 physicians and 1000 nurses
109,190 Emergency Department (ED) visits in 2010-11
24,100 admissions
23,900 surgeries
3,700 births
over 189,000 clinic visits
Has a corporate performance mgmt framework and corporate scorecard
In 2010-11, RVHS launched two transformative IT initiatives to create a competency center in business process management, and
develop an enterprise Business Intelligence system
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 31
Business Intelligence Vision at RVHS
Business Need
RVHS generates/collects a wealth of data which contain revealing facts about the quality and efficiency of RVHS’s processes, utilization of resources and outcomes.
RVHS aims to gain insights into operating performance in order to improve efficiencies and the quality of patient care.
Managers need timely access to synchronized data from all levels.
The BI system needs to provide an integrated repository of data from disparate sources organized to meet the needs of business and clinical users.
The BI system must provide a data access interface that will enable business users and to access information on their own.
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 32
Business Problem: Emergency Department Patient Flow
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 33
Arrival Triage Registration
Arrival in ED
Non Physician
Initial Assessment
(NPIA)
Special Consult
Clinical Decision
Unit(CDU)
Physician Initial
Assessment (PIA)
Treatment in ED
Disposition Decision
Patient Left ED
Departure from ED
ED Length of Stay
The Emergency Room National Ambulatory Initiative (ERNI) measures and reports how long patients spend in Emergency Departments. Clinicians
(will) collect 38 data elements (DART) related to the patient journey through the Emergency Department from arrival to departure.
Improve the quality of Patient care
June 2012 24th International Conference on Advanced Information Systems Engineering
(CAiSE'12)
34
The Seven Phases for the Design of the Emergency Department Data Mart: A Mixed approach
Process WorkflowDesign
Goal/Strategy
Map Design
Indicator Map
Design
Process Map
Design
Actor Map Design
Resource Map
Design
Goal IndicatorObject Graph
Goa/StrategyMap
ProcessWorkflow
ProcessMap
IndicatorMap
ResourceMap
ActorMap
BIM Requirement Specification
Indicator RequirementDefinement
Goal Indicator Object GraphDefinement
----------------------------------------
IndicatorTemplate
Goal IndicatorObject Graph
ANALYSIS AND RECONCILIATION
STAGINGDESIGN
CONCEPTUAL DESIGN
REQUIREMENT ANALYSIS
----------------------------------------
Preliminaryworkload
----------------------------------------
Workloaddata volume
Logicalschema
-------------
--------
--------
---
--------
---
Factschema
Requirementspecification
----------------------------------------
ETLprocedures
LOGICALDESIGN
PHYSICALDESIGN
WORKLOADREFINEMENT
-------------
--------
--------
---
--------
---
----------------------------------------
Operationalsource
schemata
Reconciledschema
Physicalschema
Parallel ActivitiesThe outputs of one activity
are used as inputsfor the other activity in
a continue loop-cycle refinement
DBMS
XMLRelationaletc.
LogicalModel
------------------------------------
Strategy goalsDocument
Users Requirements
Data Warehouse Design: Modern Principles and
Methodologies, Matteo Golfarelli and Stefano Rizzi, (2009)
Requirement Analysis with
BIM
Requirement Analysis: Actor Goal Indicator Object (AGIO)
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 35
Dart Dart • 38 Indicators
AGIO
Sheet
AGIO
Sheet • Informal
Requirement
AGIO
Graph
AGIO
Graph • Formal
Requirement
• General description • Organization's Context • Measurement • Data Mart and Navigability • Performance Parameters • Data sources details • Security / Data Access • Information and Data Quality
Reduce the percentage of ER
Left Without Being Seen patients
Percentage of ED LWBS patients(ID 7)
Time
Location
Patient
<responsible for>
Physician Initial Assessment
ER Manager
<measure><evaluate>
Who
What Why
Which (perspective)
From AGIO Sheet and AGIO Graph
Extrapolate:
Actor Map
Goals/Strategy Map
Indicator Map
Process and Workflow Map
Resource Map
Whatever combination of the above:
e.g., Goal/Strategy Map + Indicator Map
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 36
Goal/ Strategy Map + Indicators
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 37
Reduce the percentage of
admitted ED patients
Percentage of ED Visits Admitted
(ID 8)
[GOAL NOT DEFINED] the total number of patient
visits
Total ED Visits(ID 1)
?
[GOAL NOT DEFINED] the percentage of
patient visits classified as CTAS I / II / III / IV / V
Percentage of Emergency Department
Visits CTAS I / II / III / IV / V
(ID 2-6)
?
Reduce the percentage of ED
LWBS patients
Percentage of ED LWBS patients(ID 7)
Reduce the LOS_ED of ED patients in the
Emergency Department
Average LOS_ED - all dispositions
(ID 9)
Average LOS_ED for non-admitted
patients(ID 10)
Reduce the LOS_ED of ED CTAS I-II non-admitted
patients to equals or less than 7 hours
Percentage of ED CTAS I-II non-admitted patients with
LOS equals or less than 7
hours(ID 11)
Reduce the LOS_ED of ED CTAS III non-admitted patients to equals or less
than 7 hours
Percentage of ED CTAS III non-admitted patients with
LOS equals or less than 7
hours(ID 12)
Reduce the LOS_ED of ED CTAS IV-V non-
admitted patients to equals or less than 4 hours
Percentage of ED CTAS IV-V non-admitted patients with
LOS equals or less than 4
hours(ID 13)
Reduce the LOS_ED of ED admitted patients
Average LOS_ED for
admitted patients(ID 14)
Reduce the LOS_ED of ED CTAS I-II
admitted patients to equals or less than 8
hours
Reduce the LOS_ED of ED CTAS III
admitted patients to equals or less than 8
hours
Reduce the LOS_ED of ED CTAS IV-V
admitted patients to equals or less than 8
hours
Percentage of ED CTAS IV-V
admitted patients with
LOS equals or less than 8
hours(ID 17)
Percentage of ED CTAS I-II
admitted patients with
LOS equals or less than 8
hours(ID 15)
Percentage of ED CTAS III
admitted patients with
LOS equals or less than 8
hours(ID 16)
<evaluate><evaluate>
<evaluate>
<evaluate><evaluate>
Reduce the LOS_ED of ED non-admitted patients
Percentage of non-admitted patients with
LOS_ED equals or less
than <a specified time>
in hours<ID ABS-1>
<evaluate>
<influence> <influence> <influence>
LEGEND
LOS = Length of StayED = Emergency DepartmentCTAS = Canadian Triage and Acuity Scale
<evaluate>
<evaluate>
<influence> <influence> <influence>
Percentage of admitted
patients with LOS_ED
equals or less than <a
specified time> in hours
<ID ABS-2>
<evaluate>
<evaluate>
<evaluate>
<influence>
<influence>
<evaluate>
Negative Indicator
Goal
Indicator Type not defined
?
<evaluate>
<evaluate>
Improve the level care of ED patients
Improve the level care of IP patients
Improve the level care of patients
<influence> <influence>
<influence> <influence> <influence><influence>
….
….
….<influence>
<influence>
<influence>
Positive Indicator
Level care of ED patients(ID ABS-3)
<evaluate>
Level care of IP patients
(ID ABS-4)
<evaluate>
Level care of patients
(ID ABS-5)
<evaluate>
<influence>
Time
Location
CTAS
Dimension
Time
Location
CTAS
Time
Location
Time
Location
Time
Location
PatientTime
Location
CTAS
Time
Location
CTAS
Provider
Time
Location
CTAS
Provider
Time
Location
Time
Location
CTAS
Provider
Time
Location
CTAS
Time
Location
CTAS
Time
Location
CTASProvider
Provider
Time
Location
CTAS
Time
Location
CTAS
Time
Location
CTAS
Provider
ProviderProvider
Time
Location
CTAS
Provider
Time
Location
CTAS
Provider
LOS Hours
LOS Hours
Provider
Level of care of the
patients
Actor Map
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 38
Senior Management
Team(SMT)
Management Information
System Clinical Team(MIS)
Clinician
Performance Evaluation
Service Team(PES)
Hospital
Middle Management
Team
Staff
ANDAND
Floor Manager
Emergency Department
Manager(ED Manager)
Unit Clerk
Support Service
AND
Registered Nurse (RN)
Physician(MD)
Nurse Practitioner
(NP)
Registered Practical Nurse
(RPN)
Therapist
AND
Physiotherapist (PT)
Occupational Therapist
(OT)
AND
Infection Control Practitioner
(ICP)
Triage
Assessm
ent
Left
ED
Registrat
ion
Decisio
n/
Disposit
ion To
Admit
Triage
DateTime
<produce>
Service
DateTime
<produce>
Task Information
Provider
InitialAsses
DateTime
<produce>
Physician
Initial
Assessm
ent
NegativeIndicator
Time to PIA
Seen
By
Consult
ant
Consult
ant
Request
<produce>
Consultant
Called
DateTime
<produce>
Consultant
Arrived
DateTime
<produce>
AdmitDecision
DateTime
<produce>
Depa
rt
Er
DateTime
<produce>
Left ER
DateTime
Waiting for
Consultant
Time to get
a Bed
LOS_ED
LEGEND
Transfer
of Care
Transfer of Care
DateTime
<produce>
STA
RT
join in
one
path
DO both
paths IF
Ambulance
Arrive =
True ELSE
DO
Registration
path only
Control Flow Node
END
START/END point
<measure>
<measure>
<measure>
<measure><measure> <measure>
<measure>
<measure>
Note: Ambulance
arrive and Patient
Walki-in sub-
processes are
hided due to
space constrains
Emergency Department Process + Indicators
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 39
eDART (daily):1) # of Visits2) % of CTAS 13) % of CTAS 24) % of CTAS 35) % of CTAS 46) % of CTAS 57) % Left Without Being Seen8) % of Visits Admitted9) AVG LOS All dispositions10) AVG LOS Non admitted patients11) % of CTAS 1-2 Non Admitted patients with LOS <= 7 hours12) % of CTAS 3 Non Admitted patients with LOS <= 7 hours13) % of CTAS 4-5 Non Admitted patients with LOS <= 7 hours14) AVG LOS Admitted patients15) % of CTAS 1-2 Admitted patients with LOS <= 8 hours16) % of CTAS 3 Admitted patients with LOS <= 8 hours17) % of CTAS 4-5 Admitted patients with LOS <= 8 hours
Dashboard (hourly):a) AVG Time to Physician Initial Assessmentb) AVG Waiting time for a Bed
ED Visit(01/09/2011 - Present)
- Triage,- Registration,- Consultation Request- Consultation Performed- Discharge,- Disposition,- Left ED
DayHoliday
YearQuarter
Week
Month
Date
Provider
CTAS
Level
Code
Diagnosis
Location
Bed
Ward
Name
Group
Description
Floor/ Unit
Faciltiy
Time
HH
MM
SS
Desciption
ED#DART:#
0
10
20
30
40
1 4 7 10 13 16 19 22
Pa*ents#in#ED#
total
#visits
AvgVis
avgTot
Emergency#Department# :#
Admissions#(since#12AM):#5 #Wai*ng#for#Consult:#8#
Wai*ng#for#bed:#2 #Wai*ng#for#DI:#3#
ED#White#Board
DART#Trending
Inpa*ent#Ac*vity ALC#Analysis
FEED
(A)
(B)
ED Fact Schema and a Dashboard
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 40
We derived it from the Indicator Map
Lessons Learnt What is the value of BIM in a BI implementation?
BIM concepts enhance communication and collaboration between designers and domain experts
Provide a roadmap for project team
Is the initial BIM language sufficient to support the business modeling needs of the case study? Used concepts such as stakeholders, goals, processes, KPIs,
scorecard, resources, etc …
Some concepts and methods not used (situations, reasoning, …)
Who are the users of BIM? Business analysts and not business managers
Designers and domain experts understood and used the models for communication
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 41
Lessons Learnt Is there a development methodology that matches with BIM?
Extended widely practiced BI solution development techniques by enriching them with BIM concepts
How does BIM map to data?
Indicator maps used to derive fact schemas, map current indicators to objectives
Future WORK: Toward a Model for Performance Management Solution
Performance management enables organizations to monitor performance across the business, linking performance to business cycles and strategies that govern their overall direction.
Use the formal requirements output of our Requirement analysis as input for Performance Management Frameworks in BI platforms to validate our BIM model as a Performance Management Solution Model
June 2012 24th International Conference on Advanced Information Systems Engineering (CAiSE'12) 42
* WebFOCUS Performance Management Framework v5, Information Builders (2009)
Conclusions Business Intelligence Model
Fill the gap between BI data and business strategy
Language and reasoning consolidation Consistent “story” or use cases for all types of BIM reasoning
Hybrid reasoning with incomplete indicators
BIM Language Semantics and Reasoning Language (re)design, formal definition, use of metaproperties,
expanded reasoning
BIM in Action
Future Work Expanded description of semantics and reasoning
Business Intelligence Model 43
Thank you!
Questions?
Contact:
Business Intelligence Model 44