Integration of ELECTRE TRI in a GIS Coupling with a XMCDA webservice - PowerPoint PPT Presentation

Quick reminder Objectives update New developments Demo What’s next ? Integration of ELECTRE TRI in a GIS Coupling with a XMCDA webservice for inference Olivier Sobrie University of Mons Faculty of engineering April 13, 2010

Quick reminder Objectives update New developments Demo What’s next ? Quick reminder 1 Objectives update 2 New developments 3 Demo 4 What’s next ? 5

Quick reminder Objectives update New developments Demo What’s next ? GIS and MCDA Combination Spatial Query GIS Visualization Organization Prediction Analysis ◮ GIS are used in lot of application from land suitability problem to geomarketing ◮ Since 90’s, works about GIS and MCDA ◮ Not a lot of work based on ELECTRE methods ◮ ELECTRE methods fit well for ordinal problems

Quick reminder Objectives update New developments Demo What’s next ? GIS and MCDA Limitations of GIS-MCDA works according to S. Chakhar : ◮ Weak coupling ◮ One MCDA method integrated ◮ Choice of the MCDA method ◮ Single criterion synthesis ◮ User’s knowledge of SIG and MCDA

Quick reminder Objectives update New developments Demo What’s next ? GIS and MCDA Limitations of GIS-MCDA works according to S. Chakhar : ◮ Weak coupling ◮ One MCDA method integrated ◮ Choice of the MCDA method ◮ Single criterion synthesis ◮ User’s knowledge of SIG and MCDA We add an extra one : A good number of GIS-MCDA tools were abandoned or never surpassed the stage of prototype

Quick reminder Objectives update New developments Demo What’s next ? Objectives of our GIS-MCDA integration ◮ ELECTRE TRI implementation ◮ Tight coupling ◮ User friendly interface ◮ Open Source GIS (and implementation) ◮ Support for standard and Bouyssou-Marchant methodology

Quick reminder Objectives update New developments Demo What’s next ? Strategy to build the decision map Criterion map 1 Criterion map 2 Criterion map 3 Step 1: Construction of criterion maps Multicriteria map Step 2: Construction of an intermediate map ELECTRE TRI Inference Step 3: ELECTRE TRI model module module Decision map Step 4: Generation of the decision map

Quick reminder Objectives update New developments Demo What’s next ? Status at the previous workshop

Quick reminder Objectives update New developments Demo What’s next ? Demo : Densification of Quebec city Subject Quebec city wants to create a program to densify its population in the centrum and around the small crown. The program consists to build rental properties at low prices for young families in empty areas. Objectives ◮ Densify central sectors where the there are more public transports ◮ Sustain a good social diversity by choosing in priority the sectors where young people and immigrants are not well represented ◮ Favor sectors with a lot of small shops

Quick reminder Objectives update New developments Demo What’s next ? Demo : Densification of Quebec city Actions 786 actions (polygons) Criteria ◮ Density of 0-14 years old [%] (min) ◮ Density of shops [shops/ha] (max) ◮ Density of people [residents/ha] (min) ◮ Level of public transports (average) [bus/hour] (max) ◮ Ratio of immigrants [%] (min) Categories 1. Bad 2. Medium 3. Good

Quick reminder Objectives update New developments Demo What’s next ? Objectives update Save/Load parameters Add the possibility to save an XMCDA model and restore it in the plugin XMCDA webservice for parameters inference ◮ Create a new webservice to infer parameters of the ELECTRE TRI model globaly and partialy ◮ Make some experiments Coupling the webservice with our ELECTRE TRI plugin Create user-friendly interface to use the webservice with our Quantum GIS plugin

Quick reminder Objectives update New developments Demo What’s next ? Save/Load parameters

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference webservice Categories profiles Learning alternatives Performance table of profiles Criteria Criteria weights XMCDA Performance table webservice Credibility threshold Categories Compatible alternatives Affectations Message Characteristics ◮ Bouyssou-Marchant ELECTRE TRI model ◮ Accept non-admissible set of learning alternatives ◮ Maximize number of compatible alternatives ◮ MIP problem ◮ Use GLPK

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations Methodology Similar methodology as the one used by Agnès Leroy in her thesis Step 1 : Generate random data Set of random Random ELECTRE TRI Sorted alternatives alternatives model C k +1 C k g n C k g n − 1 g j g 2 C k +1 g 1

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations Methodology Similar methodology as the one used by Agnès Leroy in her thesis Step 1 : Generate random data Set of random Random ELECTRE TRI Sorted alternatives alternatives model C k +1 C k g n C k g n − 1 g j g 2 C k +1 g 1 Step 2 : Pick learning alternatives Set of random alternatives Learning set

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations Methodology Step 3 : Inference of ELECTRE TRI model Set of learn- Learned ELECTRE TRI ing alternatives model C k C k +1 g n C k g n − 1 Inference g j Program g 2 C k +1 g 1

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations Methodology Step 3 : Inference of ELECTRE TRI model Set of learn- Learned ELECTRE TRI ing alternatives model C k C k +1 g n C k g n − 1 Inference g j Program g 2 C k +1 g 1 Step 4 : Analysis of learning model Alternatives sorted Original ELECTRE TRI by the original model model C k +1 C k g n C k g n − 1 g j Set of random g 2 alternatives C k +1 g 1 Alternatives sorted Learned ELECTRE TRI model by the learned model C k C k +1 g n ′ C g n − 1 k g j g 2 ′ C k +1 g 1

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations Results - Affectation errors Affectation errors for a model with 2 categories Affectation errors for a model with 4 criteria 20 50 3 criteria 2 categories 4 criteria 3 categories % of affectation errors % of affectation errors 4 categories 5 criteria 40 15 30 10 20 5 10 0 10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100 Number of learning alternatives Number of learning alternatives Remarks ◮ Number of criteria ր ⇒ Affectation error ր ◮ Number of categories ր Affectation error ր ⇒ ◮ Number of learning alt. ր ⇒ Affectation error ց

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations Results - Computing time Computing time for a model with 2 categories Computing time for a model with 4 criteria 1 , 200 80 3 criteria 2 categories 4 criteria 3 categories 1 , 000 Computing time (secs) Computing time (secs) 5 criteria 4 categories 60 800 600 40 400 20 200 0 0 10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100 Number of learning alternatives Number of learning alternatives Remarks ◮ Number of criteria ր ⇒ Computing time ր ◮ Number of categories ր Computing time ր ⇒ ◮ Number of learning alt. ր ⇒ Computing time ր ր

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations Results - Influence of errors in learning set Affectation errors for a model with 2 categories and 4 criteria Percentage of erroned learning alternatives rejected % of erroned learning alternives rejected 100 30 No affectation errors 10% of affectation errors 10% of affectation errors 20% of affectation errors 80 % of affectation errors 20% of affectation errors 20 60 40 10 20 0 10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100 Number of learning alternatives Number of learning alternatives Remarks ◮ Number of erroned learn. alt. ր Affectation errors ր ⇒ ◮ Number of learning alt. ր ⇒ Affectation errors ց ◮ Number of learning alt. ր Err. learn. alt. rej. ր ⇒

Quick reminder Objectives update New developments Demo What’s next ? ELECTRE TRI BM inference experimentations First conclusions and ideas for improvement First conclusions ◮ Lot of learning alternatives needed to get good results ◮ With errors in the learning set, more alternatives are needed ◮ Computing become huge when number of learning alternatives increase Ideas for improvement ◮ Two step inference ◮ Improve objective of the inference program ◮ Partial inference