In this paper, we present a solution developed at BT's Intelligent Systems Research Centre which addresses the issue of proactive service planning with the ultimate goal of aligning supply optimally to meet the anticipated demand. The planning process generates two plans; a capacity plan and a deployment plan. The capacity plan provides all relevant details in volume in terms of the anticipated demand and the available supply whereas the deployment plan is a finer-grained refinement of the capacity plan which specifies where resources should be deployed (geographical locations) and what resources should be used for (skill assignments). These plans are generated by efficiently matching a pool of available resources to a number of jobs that need to be done using an advanced search algorithm. Our optimisation approach incorporates a number of rules and parameters in order to satisfy variable sets of goals, for example to minimise cost, to maximise quality of service, or combinations of both
[1]
J. K. Lenstra,et al.
Local Search in Combinatorial Optimisation.
,
1997
.
[2]
C. D. Gelatt,et al.
Optimization by Simulated Annealing
,
1983,
Science.
[3]
John H. Holland,et al.
Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
,
1992
.
[4]
Gilbert Owusu,et al.
ARMS: An automated resource management system for British Telecommunications plc
,
2006,
Eur. J. Oper. Res..
[5]
C. Voudouris,et al.
ARMS — application of AI and OR methods to resource management
,
2003
.
[6]
Fred Glover,et al.
Tabu Search - Part II
,
1989,
INFORMS J. Comput..
[7]
F. Glover,et al.
In Modern Heuristic Techniques for Combinatorial Problems
,
1993
.
[8]
Fred W. Glover,et al.
Tabu Search - Part I
,
1989,
INFORMS J. Comput..