forecasting and scenario planning the challanges of uncertainty and complexity
TRANSCRIPT
Forecasting and Scenario PlanningThe Challanges of Uncertainty and Complexity
Agenda
Introduction on forecasting
Introduction
Methods of forecasting depend on two variables
Uncertainty
The degree of available knowledge about the target variable
Complexity
The number of variables and the extent to which they are interrelated
Introduction
Low uncertainty / Low complexitt (LL)
Point estimates
A single number
Forecasting the demand of a known product in a stable market
Introduction
High uncertainty / Low complexity (HL)
Confidence ranges
Upper and lower boundaries
Estimate the oil reserves in a new offshore field
Introduction
Low uncertainty / High complexity (LH)
Deterministic models
Identify, measure and relate different influencing variables
Fundemental forecasting of exchange rates
Introduction
High uncertainty / High complexity (HH)
Scenario planning
Introduction
Deterministic models
Scenario planning
Point estimates
Confidence ranges
High complexity
Lowcomplexity
LowUncertainty
HighUncertainty
The Challange of Uncertainty
Definition of uncertainty
As to the correct value of an unknown quantity of interest:
disagreement among forecasters
doubts within a single forecaster
The Challange of Uncertainty
Why disagreement?
Overconfidence (marriage, business success)
Illusion of control (choose lottery number)
Information distortion (availability heuristic)
Risk perception
Assessing or weighing probabilities
The Challenge of Complexity
Definition of complexity
How many variables
How deeply they interact
The Challanges of Complexity
Cognitive simplifications
The abbility to identify all variables and combine them sufficiently
Bounded rationality
Cognitive devices (for instance schema’s and associateive networks)
The Challanges of Complexity
Dynamic complexities
Cross-sectional complexities
How variables interrelate at a given point in time
Dynamic complexity
How variables interrelate over time
Research shows that people have a difficulty learning over time => internalisation of the concept not to give up and try and try again.
The Challenges of Complexity
“We should make things as simple as possible,
but not simpler” (Albert Einstein)
Uncertainty / Complexity
LL (Point estimates)
Problems
Focus extensively on one value
Reluctant to diverge from an initial estimate
Use of heuristics and consequent bias
Improve
Extrapolating forecasting
NPV analysis
Decision analytic models
Uncertainty / Complexity
HL (Confidence ranges)
Problems
Overconfidence which makes ranges to narrow
Ignoring base rates
Ignoring sample size
Improve
Event diagrams
Fault trees
Simulation
Uncertainty / Complexity
LH (Deterministic models)
Problem
Humans simplistic notion about cause and effect
Think about everything that is said during election campaigns
Improve
Assumption analysis
Influence diagrams
Uncertainty / Complexity
HH (Scenario Planning)
Problem
Inability of human beings to deal with ambiguity
Improve
Focus on exploration and learning
Feedback loops
Posing disconferming questions
Scenario Planning
Focus on the joint effect of many factors
Helps us understand how various strands of a complex tapestry move if on or more threads are pulled
The art of scenario planning lies in blending the known and the unknown into a limited number of internally consistent views of the future than span a very wide range of possibilities.
How to develop scenarios
Define the scope and time frame
Identify the major stakeholders Overview of players, there interests and power position
Identify basic trends DESTEP
Why is it relevant
Direction of the trend (positive, neuteral, negative)
How to develop scenarios
Identify key uncertainties Aimed at issues concerning the company
Determine possible outcomes for each uncertainty
Construct initial scenario themes Identify two extreme worlds by putting all the positive
outcomes in one and all the negative outcomes in the other scenario
Or, select the two most significant uncertainties and postulate two rather extreme outcomes for each uncertainty
How to develop scenarios
Build a scenario blueprint Create a matrix that represents the identified uncertainties and plot
the scenarios you constructed
Check whether all scenarios are:
Plausible
Internally consistent
Sufficiently relevant
Develop full-fledged scenarios Identify general themes
The overall goal is to identify storylines and themes for the scenarios that cover a full range of possible outcomes for the uncertain issues at hand
Highlight competing perspective
How to develop scenarios
Identify research needs To flesh out your understanding of uncertainties and
trends, as well as perhaps their interrelationship over time
Develop influence diagrams
Develop quantitative models Simulations
Usually done with specific software
How to develop scenarios
Evolve towards decision scenarios Converge towards a limited number of distinctive different
scenarios that you will eventually use to test your strategies and generate new ideas
Windtunnel the scenarios with the business ides
Assignment / Break‘Your personal scenario’s’
Take 20 minutes to answer the following questions: Which two variables in your environment will be most
important in the way your life will develop over the next 5 – 10 years?
For both variables, determine (realistically) what the most negative an most positive outcome would be for you.
Combine the 4 outcomes => 4 possible futures
How well are you prepared for these 4 possible futures and why?
What do you possible have to do the be better prepared for these 4 possible futures?
After 20 minutes I would like to ask 3 -4 students to elaborate on their scenario’s.