Flexible Workforce Management (2004-)

Constraint Satisfaction and Optimization Laboratory and
Computational Finance Research Laboratory
University of Essex
Sponsor:


This project is in collaboration with BT. Although it is motivated by BT's workforce scheduling problem, the ideas developed in this project are general. It involved scheduling engineers to jobs, satisfying a wide range of constraints. This is a multi-objective optimization problem. Some of the objectives are to minimize travelling distance and to maximize service quality as defined by the company.

Staff empowerment is also a major theme in this project. The concept of staff empowerment is to produce all-win solutions to enterprises. If properly implemented, staff will gain in job satisfaction. As a result, the enterprise will enjoy improved morale and improved productivity.

Applying staff empowerment to BT's workforce scheduling, the objective is to define bargaining mechanisms (which includes communication protocols) to enable various service regions in BT to schedule their workforce efficiently. By employing a market mechanism, we aim to help the management and the region managers to generate all-win solutions.

A retractable contract network protocol (RECONNET) has been defined. This protocol enables the system to conduct local search methods to search for near-optimal solutions. Guided Local Search is one of those meta-heuristic methods under consideration due to its simplicity and success elsewhere. As the management has multiple objectives, this is a multi-objective optimization problem, which is also a major research area in the group.

People People icon University of Essex and British Telecom
Papers Publications icon Overview paper: E.P.K. Tsang, B. Virginas, T. Gosling, W. Liu, Multi-agent based scheduling for staff empowerment, Chapter 20, in C. Voudouris, G. Owusu, R. Dorne & D. Lesaint (ed.), Service Chain Management, Springer, 2008, 263-272 (early version)
Related Projects Projects icon Modelling, Simulation and machine learning / Guied Local Search