Development of a GIS-based Expert System for on-site Storm-water Management

Urbanization is accelerating worldwide. One of the negative effects of urbanization is the overloading of the city sewer system. To solve this problem, on-site storm water infiltration proves very promising due to its near natural characteristics and multiple effects on the drainage of stormwater runoff in urban areas. However, the judgment of whether a local area is appropriate to be drained in this way and which infiltration measures are optimal is rather complex and involves analysing a set of influential factors. This judgment depends on not only relevant theoretical considerations, but also a large amount of practical experience and the availability of relevant data, as well. Such a judgment is an unstructured problem and relates to changeable knowledge. To fulfill this task, the so-called expert system, or knowledge-based system, is introduced. One of the advantages of an expert system is that it provides automation of expert-level judgment. This is extremely helpful when an expert-level judgment is needed repeatedly for a large amount of cases, like in the planning of on-site stormwater infiltration systems for an entire city catchment. This paper describes a self-developed expert system tool for developing rule-based expert systems, as well as a case study: using an expert system for the selection of on-site storm water infiltration measures for the city of Chemnitz, Germany.