Using Agents to Model Logistics
Abstract
Many areas of military concern, such as command and control of forces, operations on urban terrain, and humanitarian assistance/disaster relief operations, are replete with human interactions. The complexity of these operations along with the confluence of possible interactions outside of the control of the local commander creates situations that are not easily modeled. Increasingly these multifaceted operations are being studied using agent-based models (ABMs). While ABMs have been around for some time, those used in the military have been war fighting models. The agents possess weapons, abilities to sense, communicate, and move, and different allegiances. However, they do not have the abilities to carry, consume, or transfer resources. We describe the minimal core of capability for ABMs to successfully support logistics operations: defining a commodity, attaching it to an agent, transferring it between agents, and defining appropriate measures. Logistics functions are extremely important for the U.S. Armed Forces, and logisticians must be prepared to take the lead in drafting operational plans. The ability to quickly put together ABMs and explore their behavior over a wide range of parameter settings might help support logistics operations. We illustrate the process for logistics support model in an urban, humanitarian assistance/disaster relief scenario.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 31, 2003
- Accession Number
- ADA418397
Entities
People
- Eric S. Wolf
- Lloyd P. Brown
- Niki C. Goerger
- Susan M. Sanchez
Organizations
- Naval Postgraduate School