Slides: 1 - Spatial Complexity and MCDA Introduction.pdf


Spatial complexity

A problem that scales in complexity with multiple dimensions

Influenced by several criteria, with different importance

Often multiple decision-makers, with different/conflicting preferences

A classic example of an SDSS problem is location allocation: choosing locations such that we can optimally provide facilities.

Spatial analysis

GIS

A system for spatial analysis Geographic information system - Wikipedia

Two approaches to modeling

  • Statistical modeling
  • Mathematical modeling
    • Simulation
    • Optimization

Spatial Simulation

  • System components are mathematically defined
  • Components are related to each other in a series of functional relationships
  • Result: a mathematical description of a decision process
    • model is solved repeatedly, using different parameters and different decision variables every time
    • As these values are changed, a range of solutions are obtained for the problem and the ā€˜best’ solution can be chosen from that range.

Decision variables are the variables in a scenario that you, as a designer, have control over.

Spatial Optimization

In Spatial Optimization, we often try to maximize/minimize some quantity, which is the result of some decision variables.

This quantity is often referred to as the objective function or criterion function.

Alternatively, find some feasible alternative.

Simulation vs Optimization

Simulation modeling starts with actions and studies the effect.

Optimization starts with a definition of system objectives, and specify the actions that will best satisfy those objectives.

Semi-Structured Decisions

Structured decisions are relatively straight-forward. They can usually be optimized by optimizing one mathematical formula. There are no conflicting interests.

Unstructured decisions don’t have an optimal solution.

Semi-structured decisions fall somewhere inbetween or have aspects of both.

The structured part of the semi-structured problem may be amenable to automated solution by the use of a computer, while the unstructured aspects are tackled by decision makers.

Spatial Decision Support System

SDSS is meant to improve the effectiveness of decision making by incorporation decision makers’ knowledge and experience into computer-based procedures.

Spatial MCDA

There are three GIS-MCDA methods

Conventional MCDA

  • Mainly aspatial

Spatially Explicit MCDA

If:

  • Outcomes vary when feasible alternatives are relocated
  • Decision alternatives are spatially defined

Spatial multicriteria optimization

Specifically designed for modeling spatial systems and solving spatial problems such as:

  • Site search
  • Location allocation
  • Routing

TODO

Introduction assignments on brightspace!