Resource management problems are problems in which the goal is to optimize the utilization of resources over time. Dependencies between resources make it necessary to plan the proper utilization of these resources. Such a planning must often be constructed in an environment that is non-deterministic, for example due to competing agents. Traditional state-based planning approaches are not suited for this type of planning problems. This paper addresses an important subclass of resource management problems, namely production management problems. In this class of problems, the production of a set of goal products must be optimized over time, in the absence of non-determinism and competing agents. Three methods for production management problems are investigated, viz., an extensive search method and two abductive planners using greedy search. One abductive planner uses a heuristic that has been designed especially for production management problems while the other abductive planner learns a heuristic from experience. Experiments performed on a large and varied set of production management problems show that extensive search cannot be used for production management problems and that the abductive planner using the learned heuristic significantly outperforms the abductive planner using the designed heuristic. This indicates that the integration of learning elements in planning algorithms can improve the performance and versatility of these algorithms.
[1]
Nico Roos,et al.
Requirements for Resource Management Game AI
,
2005,
IJCAI 2005.
[2]
M. Veloso,et al.
OBDD-Based Optimistic and Strong Cyclic Adversarial Planning
,
2014
.
[3]
Clark C. Abt,et al.
Serious games
,
2016,
Springer International Publishing.
[4]
Alexander Nareyek.
Beyond the Plan-Length Criterion
,
2000,
Local Search for Planning and Scheduling.
[5]
M. M. De Weerdt,et al.
Plan Merging in Multi-Agent Systems
,
2003
.
[6]
Bernhard Nebel,et al.
The FF Planning System: Fast Plan Generation Through Heuristic Search
,
2011,
J. Artif. Intell. Res..
[7]
S.J.J. Smith,et al.
Empirical Methods for Artificial Intelligence
,
1995
.
[8]
Peter Norvig,et al.
Artificial Intelligence: A Modern Approach
,
1995
.
[9]
Tom Bylander,et al.
Complexity Results for Planning
,
1991,
IJCAI.