A Knowledge-Based System for Load Planning Hazardous Cargo.
Abstract
Military Airlift Command (MAC) must have trained personnel to accurately plan the cargo loads for its many airlift missions. These load planners are responsible for selecting the cargo for the mission, making sure that it can safely be loaded, and determining where the cargo is to be positioned. They do well under normal conditions; however, they might make mistakes when trying to keep up with the workload during a crisis. Expert system technology was investigated as a possible way to offload some of this extra work, reducing the number of errors. The goal of this report was to determine the feasibility of an expert system for doing one phase of the planning -- load planning hazardous cargo. A rule-based prototype was developed using Personal Consultant Plus to select the hazardous cargo and determine if it was safe to load. With AFR 71-4, Table A-1; the associated cargo notes; and load planning heuristics, the prototype was able to determine if the hazardous cargo could go on the build-up pallet, if it needed to be loaded loose, or if it had to wait for another flight. The report concluded that an expert system was feasible for load planning hazardous cargo. It also listed enhancements and further research that would extend the prototype and enable it to be used by load planners. Keywords: Theses; Artificial intelligence. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1986
- Accession Number
- ADA178184
Entities
People
- Roderick S. Sanborn
Organizations
- Air Force Institute of Technology