1st International Congress on Environmental Modelling and Software - Lugano, Switzerland - June 2002
Keywords
land use planning, multi-objective, genetic algorithms, soft-systems
Start Date
1-7-2002 12:00 AM
Abstract
Land managers are increasingly faced with complex decisions requiring the consideration of tradeoffs between multiple, non-commensurable objectives. Such decisions have profound effects on the financial, social and environmental sustainability of land use systems. One approach to assisting land managers with these decisions has been the development of computer-based decision support systems (DSS). While such systems are demonstrably able to make analyses of multi-objective land-use planning problems, are the answers they produce relevant and useful to practitioners? This paper reports on a workshop-based, soft-systems analysis of outputs from a spatial, multi-objective land-use planning tool. The paper outlines the approach taken in developing the decision support system, focusing on the land-use planning tools. These tools use multi-objective genetic algorithms to define the structure of the trade-off between objectives. The paper then details the softsystems- based evaluation strategy. Land managers and other professionals from a range of backgrounds were asked to devise individual “best compromise” plans, balancing financial and landscape diversity goals, for a farm in upland Scotland. Sub-groups of land managers were then set the task of agreeing on a plan between the members of the group. This process used the soft-systems methods of facilitated discussion and reporting back from sub-groups. The land managers’ and sub-groups’ plans were analysed with the DSS tools and the results compared with outputs from the land-use planning tools. From this analysis and the qualitative responses within the workshop it was possible to conclude that the land-use planning tools provided a useful means of exploring the patterns of land use that could be adopted for a land management unit. The process identified a number of assumptions made by land managers that could usefully be incorporated into the operation of the DSS. The use of soft-systems based analysis of land-use planning tool outputs is recommended, not only for evaluating the performance of the tools, but also for ensuring that the DSS is answering a correctly formulated problem.
Using soft-systems methods to evaluate the outputs from multi-objective land use planning tools
Land managers are increasingly faced with complex decisions requiring the consideration of tradeoffs between multiple, non-commensurable objectives. Such decisions have profound effects on the financial, social and environmental sustainability of land use systems. One approach to assisting land managers with these decisions has been the development of computer-based decision support systems (DSS). While such systems are demonstrably able to make analyses of multi-objective land-use planning problems, are the answers they produce relevant and useful to practitioners? This paper reports on a workshop-based, soft-systems analysis of outputs from a spatial, multi-objective land-use planning tool. The paper outlines the approach taken in developing the decision support system, focusing on the land-use planning tools. These tools use multi-objective genetic algorithms to define the structure of the trade-off between objectives. The paper then details the softsystems- based evaluation strategy. Land managers and other professionals from a range of backgrounds were asked to devise individual “best compromise” plans, balancing financial and landscape diversity goals, for a farm in upland Scotland. Sub-groups of land managers were then set the task of agreeing on a plan between the members of the group. This process used the soft-systems methods of facilitated discussion and reporting back from sub-groups. The land managers’ and sub-groups’ plans were analysed with the DSS tools and the results compared with outputs from the land-use planning tools. From this analysis and the qualitative responses within the workshop it was possible to conclude that the land-use planning tools provided a useful means of exploring the patterns of land use that could be adopted for a land management unit. The process identified a number of assumptions made by land managers that could usefully be incorporated into the operation of the DSS. The use of soft-systems based analysis of land-use planning tool outputs is recommended, not only for evaluating the performance of the tools, but also for ensuring that the DSS is answering a correctly formulated problem.
