Automated route planning for large vehicles on industrial construction sites

This dissertation presents the work done to automate the tasks required to plan access routes for large vehicles on industrial construction sites. Planning access routes within the site was identified to be an important consideration in the development of the project execution plan. Many important project management decisions such as site layout, construction sequence, facility design, and the extent of preassembly are influenced by the need for access. In this research, a structured approach to planning access routes is defined, the limitations of the current methods are identified and the requirements of an ideal planning tool are determined. Concepts from the areas of information representation, geometric reasoning, and artificial intelligence are found to be applicable for route planning. The route planning system was developed using the expert system Nexpert Object and the geographic information system Arc/Info in conjunction with the CAD package Microstation, the spreadsheet Excel, and custom programs. The primary features of the system are: (i) integration of schedule and design data with site geometry, (ii) automatic detection and display of obstacles on roads and turns, and (iii) automatic selection and display of the most suitable and alternate access routes. A route planning problem was used to test the system for functionality, solution quality, and utility. The primary conclusions made from this research are (i) the Expert GIS approach can be used to model and generate solutions to the access planning problem, (ii) Expert GIS is a powerful technology which can be applied to automate a variety of construction planning tasks.