Knowledge-Enabled Decision Support System for Routing Urban Utilities

Abstract

This paper presents a Web-based system for supporting the selection of the most suitable routes for buried urban utilities. The aim of the proposed system is to support (not make) decisions through a collaborative semiautomated environment, in which stakeholders can share information and/or study the impacts of different routing alternatives with respect to decision constraints. First, the knowledge relating to route selection for urban utilities is represented through an ontology. The ontology defines the types and attributes of infrastructure products and the surrounding areas. It also defines the impacts of routing options on surrounding areas through a set of decision criteria adopted to evaluate the effectiveness of any route in terms of its potential impacts. A set of constraints are also defined to help represent/study the decision criteria. Second, a GIS-based system has been created to help visualize route data, interact with users, and support the needed discussions among stakeholders. The portal also achieves data interoperability through wrapping existing geospatial data with ontology structures. Finally, a set of reasoners have been created to help quantify/augment some of the constraints. The system is capable of (1) extracting the attributes of each routing option, (2) testing the interaction/conflicts between route attributes and the constraints of the surrounding area, (3) studying the impacts of a route as stipulated in the ontology, (4) referring users to existing best practices to help enhance routes or address conflicts and, when needed, (5) develop objective measures for comparing different routes. On the microlevel (street level), route options are evaluated through a “constraint-satisfaction” approach. On the macrolevel (city level), route options are evaluated through a fuzzy inference scoring system. The proposed system focuses on facility life cycle, sustainability, and community impacts. Construction costing, scheduling, labor, and equipment along with other management issues can either be added to the system or, better, analyzed through integrating the system with four-dimensional (4D) modeling tools.