Public Procurement Evaluation by Evidence-based Multiple Criteria Decision Analysis — From conventional scoring to systematic profiling Professor Jian-Bo.

Slides:



Advertisements
Similar presentations
Value for Money – new requirements and challenges
Advertisements

DECISION MODELING WITH Multi-Objective Decision Making
Multi-Attribute Utility Theory (MAUT)
Multi‑Criteria Decision Making
1 Evaluation Rong Jin. 2 Evaluation  Evaluation is key to building effective and efficient search engines usually carried out in controlled experiments.
Topic 2. DECISION-MAKING TOOLS
Chapter 10 Decision Making © 2013 by Nelson Education.
 How to infer causation: 8 strategies?  How to put them together? S519.
Ecological Economics Lecture 13 Ricardo da Silva Vieira Researcher/Consultant Tiago Domingos Assistant Professor Environment and Energy Section Department.
September 20 th, 2005 Introduction to Expert Choice National Institutes of Health Office of Research Services Office of Quality Management 1.
Decision-Making Understand the main steps involved in rational decision-making Discuss the major reasons for poor decisions, and describe what managers.
Introduction to Management Science
Evaluating the Performance of Salespeople
Copyright © 2006 Pearson Education Canada Inc Course Arrangement !!! Nov. 22,Tuesday Last Class Nov. 23,WednesdayQuiz 5 Nov. 25, FridayTutorial 5.
Multi Criteria Decision Modeling Preference Ranking The Analytical Hierarchy Process.
Copyright © 2011 Pearson Prentice Hall. All rights reserved. Chapter 10 Capital Markets and the Pricing of Risk.
Introduction to Management Science
Using ranking and DCE data to value health states on the QALY scale using conventional and Bayesian methods Theresa Cain.
1 Decision Analysis by Dr. AA. 2 Man decides based on what he believes… Man believes what he want to believe…
Scaling and Attitude Measurement in Travel and Hospitality Research Research Methodologies CHAPTER 11.
1 Copyright © 2000 by Harcourt, Inc. All rights reserved. (1) 11 Evaluating the Performance of Salespeople Module 11 Evaluating the Performance of Salespeople.
«Enhance of ship safety based on maintenance strategies by applying of Analytic Hierarchy Process» DAGKINIS IOANNIS, Dr. NIKITAKOS NIKITAS University of.
Presented by Johanna Lind and Anna Schurba Facility Location Planning using the Analytic Hierarchy Process Specialisation Seminar „Facility Location Planning“
4 MARKETING STRATEGY O.C. FERRELL • MICHAEL D. HARTLINE SWOT Analysis
Vision & Mission Strategy Formulation External Opportunities & Threats Internal Strengths & Weaknesses Long-Term Objectives Alternative Strategies Strategy.
ELearning / MCDA Systems Analysis Laboratory Helsinki University of Technology Introduction to Value Tree Analysis eLearning resources / MCDA team Director.
Some Background Assumptions Markowitz Portfolio Theory
1 1 Slide © 2004 Thomson/South-Western Chapter 17 Multicriteria Decisions n Goal Programming n Goal Programming: Formulation and Graphical Solution and.
A decision making model for management executive planned behaviour in higher education by Laurentiu David M.Sc.Eng., M.Eng., M.B.A. Doctoral student at.
Roles of Economists and New Analytical Requirements
Spreadsheet Modeling and Decision Analysis, 3e, by Cliff Ragsdale. © 2001 South-Western/Thomson Learning Multicriteria Decision Making u Decision.
Lesson 8: Effectiveness Macerata, 11 December Alessandro Valenza, Director, t33 srl.
Probability & Statistics – Bell Ringer  Make a list of all the possible places where you encounter probability or statistics in your everyday life. 1.
THE ANALYTIC HIERARCHY PROCESS EXTENSIONS. AHP VALIDATION EXERCISE This exercise helps to validate the AHP. You will make judgments on the relative sizes.
Multi-Criteria Decision Making by: Mehrdad ghafoori Saber seyyed ali
Chapter 9 - Multicriteria Decision Making 1 Chapter 9 Multicriteria Decision Making Introduction to Management Science 8th Edition by Bernard W. Taylor.
Investment Analysis and Portfolio Management First Canadian Edition By Reilly, Brown, Hedges, Chang 6.
The Nature and Method of Economics 1 C H A P T E R.
Chapter McGraw-Hill/Irwin Copyright © 2008 by The McGraw-Hill Companies, Inc. All rights reserved. Risk and Capital Budgeting 13.
An overview of multi-criteria analysis techniques The main role of the techniques is to deal with the difficulties that human decision-makers have been.
Learning Objectives Copyright © 2002 South-Western/Thomson Learning Using Measurement Scales to Build Marketing Effectiveness CHAPTER ten.
BUSINESS PERFORMANCE MANAGEMENT
Uncertainty Management in Rule-based Expert Systems
Preference Modelling and Decision Support Roman Słowiński Poznań University of Technology, Poland  Roman Słowiński.
Analytic Hierarchy Process (AHP)
Open ECBCheck Methods for Quality Development Rafael García Rodríguez University of Augsburg, 2010.
Preliminary Analysis of Alternatives for the Long Term Management of Mercury John Vierow Science Applications International Corp. Reston, VA May 1, 2002.
Location Planning and Analysis Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent.
EQNet Implementation Experience in Lithuania Dr. Eugenijus Kurilovas, Virginija Bireniene, ITC MoE Lithuania EdRene / eQNet, Lisbon,
16469 Low Energy Building Design Conflict and Interaction in Environmental Engineering Design.
QUANTITATIVE TECHNIQUES
S ystems Analysis Laboratory Helsinki University of Technology 1 Decision Analysis Raimo P. Hämäläinen Systems Analysis Laboratory Helsinki University.
Applied Mathematics 1 Applications of the Multi-Weighted Scoring Model and the Analytical Hierarchy Process for the Appraisal and Evaluation of Suppliers.
Multi-Attribute Decision Making MADM Many decisions involve consideration of multiple attributes Another term: multiple criteria Examples: –Purchasing.
Preference Modelling and Decision Support Roman Słowiński Poznań University of Technology, Poland  Roman Słowiński.
Water Use Planning Siobhan Jackson BC Hydro Generation November 3, 2004 CEATI Water Management Workshop, Vancouver BC Translating Sustainability Theory.
Anne Lythgoe April What I want to do… Agree the scope of ‘social value’ Discuss why social value is important to commissioners of services and how.
ESTIMATING WEIGHT Course: Special Topics in Remote Sensing & GIS Mirza Muhammad Waqar Contact: EXT:2257 RG712.
DADSS Multiattribute Utility Theory. Administrative Details Homework Assignment 6 is due Monday. (slightly shorter) Homework Assignment 7 posted tonight.
QUANTITATIVE TECHNIQUES
Analysis Manager Training Module
6 Selecting Employees and Placing Them in Jobs
Bidding Strategies.
Regression Analysis Module 3.
Reality of Highway Construction Equipment in Palestine
Supplement S7 Supplier Selection.
Analytic Hierarchy Process (AHP)
A Scoring Model for Job Selection
Evaluating the Performance of Salespeople
Chapter 12 Analyzing Semistructured Decision Support Systems
Presentation transcript:

Public Procurement Evaluation by Evidence-based Multiple Criteria Decision Analysis — From conventional scoring to systematic profiling Professor Jian-Bo Yang Director of Decision and Cognitive Sciences Research Centre Manchester Business School (MBS) The University of Manchester Tel: (Ext: 63427), (O2) Web: Public Procurement Evaluation and MCDA by J B Yang of MBS

Outline of This Presentation  Public procurement evaluation and Multiple Criteria Decision Analysis (MCDA)  Typical MCDA models –Decision matrix – Scoring –Pairwise comparison decision matrix - Rating –Belief decision matrix – Profiling or grading  Evaluation aggregation based on scores –Linear aggregation or weighted sum –Reference point approach using nonlinear distance measures  Evaluation aggregation based on beliefs (IDS) –Evidence collection, mapping and grading –Evidential reasoning for generating bidder profiles –Expected utility as score for ranking –Sensitivity analysis for testing the robustness of ranking –Communication based on both bidder profiles and scores Public Procurement Evaluation and MCDA by J B Yang of MBS

Public Procurement Evaluation and MCDA Procurement evaluation criteria (weight)  Contractor’s organisation (0.1)  Financial considerations (0.3)  Management resources (0.2)  Past experience (0.2)  Past performance (0.2) –Failure of a contract (0.25) –Overruns: time (0.25) –Overruns: cost (0.25) –Actual quality achieved (0.25) –…… Public Procurement Evaluation and MCDA by J B Yang of MBS

Multiple Criteria Decision Analysis Under uncertainty – Summary of main features  A hierarchy of performance or risk criteria  Quantitative and qualitative criteria  Precise data and uncertain numbers  Subjective judgements with uncertainty  Possible absence of data  Non-commensurability among criteria  Conflict among criteria  Ranking may not be precise Public Procurement Evaluation and MCDA by J B Yang of MBS

Modelling for Procurement Evaluation  Transparency and fairness via knowledge sharing  Objectivity via data collection and management  Systematic analysis via information aggregation  Panoramic view of bidder profile  Sensitivity analysis for uncertainty clarification  Consistency in evaluation  Simulation for improvement and feedback  Communication with evidence (original / aggregated) Public Procurement Evaluation and MCDA by J B Yang of MBS

MCDA Models for Procurement Evaluation Scoring based model – decision matrix MCDA problem with numbers: Decision maker is faced with assessing and ranking several alternatives with all attributes being considered simultaneously, with no attribute being absolutely more important than others. The problem can be represented as follows Criterion 1Criterion 2…Criterion m Bidder 1y 11 y 12 …y1my1m Bidder 2y 21 y 22 …y1my1m …………… Bidder lyl1yl1 yl2yl2 …y lm Decision Matrix (Table) How should the assessment and ranking be made ? Public Procurement Evaluation and MCDA by J B Yang of MBS

Scoring-based Decision Matrix – Job evaluation Decision Matrix for Job Evaluation CriteriaJob offer 1Job offer 2Job offer 3 Salary£32,500£28,500£26,000 Quality of life Average (50%) Poor (25%) Good (75%) Interest of work Poor (25%) Good (75%) Average (50%) Location Poor (25%) Good (75%) Public Procurement Evaluation and MCDA by J B Yang of MBS

Pairwise Comparison Matrix Compare each pair of job offers on a criterion Pairwise Comparison Matrix for Job Evaluation Quality of life Job offer 1Job offer 2Job offer 3 Job offer Job offer Job offer 3241 Job 1 is judged (rated) twice as good as Job 2 in terms of “Quality of life” (Interval comparison ?) Public Procurement Evaluation and MCDA by J B Yang of MBS

Evidence-based Belief Decision Matrix – Take into account judgmental information Criterion 1Criterion 2…Criterion m Alternative 1y 11 S 12 …S1mS1m Alternative 2y 21 S 22 …S1mS1m …………… Alternative lyl1yl1 Sl2Sl2 …S lm Belief Decision Matrix MCDA problem with both numbers and judgements: Belief distribution : S ij ={(H 1, β ij1 ), (H 2, β ij2 ), ……, (H N, β ijN )} Public Procurement Evaluation and MCDA by J B Yang of MBS

House Criteria House 1 in Altrincham House 2 in Heaton House 3 in Mercy House 4 in Didsbury Location {(G, 0.5), (E, 0.5)} {(G, 0.5)} {(A, 0.2), (G, 0.8)} {(G, 0.2), (E, 0.8)} Distance (mile) Asking Price (£) 113,000110,000118,000150,000 Attractive- ness {(P, 0.05), (G, 0.35), (E, 0.60)} {(A, 0.4), (G, 0.6)} {(G, 0.3), (E, 0.7)} {(G, 0.6), (E, 0.4)} Belief Decision Matrix Assessment based on evidence collected Belief is generated from the assessment of evidence Public Procurement Evaluation and MCDA by J B Yang of MBS

Belief Decision Matrix Assessment based on evidence collected and mapped Assessing the Location of House 1 in Altrincham using the collected evidence against the agreed assessment standards (mapping) Public Procurement Evaluation and MCDA by J B Yang of MBS

Procurement Evaluation Aggregation – Weighted sum or Multiple Attribute Value Function General form of an additive (linear) value function is given by: Conditions for use of Additive MAVF: 1.Satisfaction of preferential independence among any groups of attributes. This is only a necessary condition. 2.Satisfaction of the corresponding trade-off, or Thomsen condition. 3.Interval scale property for constructing marginal value function. 4.Weights of attributes need to be assessed as scaling constants (trade-offs), or swing weights, not necessarily relative importance. 5.Linear & complete compensation among criteria without any limit. Public Procurement Evaluation and MCDA by J B Yang of MBS

Chinese Restaurant Menu : Combination of soup and main dish Attribute 1: Choose soup Attribute 2: Choose main dish Are you preferentially independent when choosing soup and main dish? MCDA – Value Measurement Theory – Preferential independence violation example SoupValue scoreMain dishValue score Mixed veg & bean curd 8Bean curd10 Egg and tomato 3 Pork with Spring Onions 7 If you are preferentially independent in choosing soup and main dish, you would ask for a main dish without considering what soup you have taken. However, is this the case for you? Would you really choose both Mixed veg & bean curd as soup and Bean curd as main dish? Public Procurement Evaluation and MCDA by J B Yang of MBS

Limitation or Bias of Additive MAVF Efficient frontier: A, B, D, E, F, G Efficient convex hull: A, E, G Additive MAVF cannot find B or F as the most preferred solution ωsvs+ωpvp=vωsvs+ωpvp=v Public Procurement Evaluation and MCDA by J B Yang of MBS

Distance-based Aggregation Ideal point models (minimax distance) Ideal point models: Set an ideal reference point and find an alternative closest to the ideal point in certain distance measure. Reference point Ideal point Set criterion weights Public Procurement Evaluation and MCDA by J B Yang of MBS

Evidential Reasoning MCDA Assessment distribution by a belief structure ER Example 1 : A qualitative assessment that the quality y q of a bidder A is assessed to be “ Good ” or “ Excellent ” by an equal number of assessors, with no assessment below “ Average ”, can be described by the following distribution S(y q (A)) ={(Bad, 0), (Average, 0), (Good, 0.5), (Excellent, 0.5)} which is termed as a belief distribution of assessment, with “ Bad ”, “ Average ”, “ Good ” and “ Excellent ” defined as “assessment grade” and 0 (0%) and 0.5 (50%) as “belief degree” (frequency to which “ Good” or “ Excellent ” is ticked by the assessors). The above distribution shows the quality profile of the bidder. Public Procurement Evaluation and MCDA by J B Yang of MBS

Bidder 1 Score 76% 6.1 Give examples of STRATEGIC Partnering, Alliances and Collaborative Working Bidders ({Best, 28%}, {Good, 51%}, {Average,17%}, {Poor, 4%}, {Worst, 0%} Bidder 2 Score 76% ({Best, 46%}, {Good, 29%}, {Average,15%}, {Poor, 3%}, {Worst, 7%} Traditionally, only scores are used ER uses both scores and belief degrees Evidential Reasoning MCDA Assessment Using ER – What’s different Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using a Decision Support System – Intelligent Decision System (IDS)  IDS is supported by the Evidential Reasoning (ER) approach  ER has been developed over a period of over 15 years  ER results from multi-discipline research - Decision Sciences - Artificial Intelligence - Statistical Analysis - Fuzzy Sets  ER addresses subjectivity and uncertainties  ER can handle heterogeneous information  ER guarantees to generate rational results  ER is gaining popularity in both academia and industry Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Advantages  Structured and natural No modification needed in IDS for procurement evaluation modelling  Flexible in modelling Model can be modify, attributes changed, added and deleted easily  Improved consistency and efficiency Through knowledge management and using an systematic evidence mapping process Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Advantages  No unnecessary assumption No need to use scores for subjective judgement No need to assume missing data  Transparent Candidates compared on any attribute at any level Weaknesses and strengths of each candidate identified  Rational, convincing and informative Examine impact of changes in any factor on decisions easily so that the decisions are made in a more rational, convincing and informative way Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Modelling Build an evaluation criteria hierarchy Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Modelling Define qualitative attribute : Number of grades can be changed Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Modelling Define grades to assess a qualitative attribute: Preference value of grades can be changed Wording of grades can be changed Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Modelling Define grade standard to assess a qualitative attribute: This can be used as guidelines to help improve consistency in assessment Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Modelling Define quantitative attribute: Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Modelling Assign weights Drag and drop to change weight Or type weight here Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Review Document Evidence classified and recorded Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Mange Knowledge Evidence examined and comments provided Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – Make assessment Evidence mapped and belief degrees assigned to grades Grade guidelines entered earlier Optional More than one grades may be selected Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – View Results Performance distribution (profile) of bidder generated Unknown element due to lack of data in Economic Test & Interview Can be any attribute in the hierarchy Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – View Results Performance scores – when there is missing data Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – View Results Compare candidates on multiple criteria – by scores Public Procurement Evaluation and MCDA by J B Yang of MBS

Assessment Using IDS – View Results Compare candidates – by performance profile Public Procurement Evaluation and MCDA by J B Yang of MBS

Other Applications - Siemens UK Supplier pre-qualification assessment Public Procurement Evaluation and MCDA by J B Yang of MBS

Other Applications  Product design and evaluation car, motorcycle, ship, aircraft, computer, …  Safety and risk assessment  Quality management  Supply chain management  Environmental management  Financial services and investment  Customer satisfaction survey  Web based survey  Data collection only – remote or onsite audit Public Procurement Evaluation and MCDA by J B Yang of MBS

Summary and Conclusions  Public procurement evaluation and multiple criteria decision analysis (MCDA)  Typical MCDA procurement evaluation models –Decision matrix – Scoring –Pairwise comparison decision matrix - Rating –Belief decision matrix – Profiling or grading  Evaluation aggregation based on scores –Linear aggregation or weighted sum –Reference point approach using distance measures  Evaluation aggregation based on beliefs (IDS) –Evidence collection and mapping or grading –Evidential reasoning for generating bidder profile –Expected utility as score for ranking –Sensitivity analysis for testing the robustness of ranking –Communication based on both bidder profile and score Public Procurement Evaluation and MCDA by J B Yang of MBS