Agents in a route planning application

Characteristics of software agents, such as intelligence, autonomy, interactivity, and mobility, have motivated the investigation of applying agent technology to a real-world vehicle routing problem. This paper presents a new approach to successfully deploy agents in a mobile nursing service with time windows, using non-explicit information and discretely imitating human negotiation processes. While keeping the economic cost of the routes down, the system must take the personal requirements and wishes of the nurses into consideration. We demonstrate how the incorporation of reciprocal feelings amongst the personnel in agents like software components can lead to a higher quality global satisfaction. The agents introduced in this approach do not take part in the pure vehicle routing part of the problem; their role is restricted to representing the personnel and negotiating the assignment of trajectories on their behalf.