As already introduced earlier, scheduling corresponds to the activity of assigning production tasks to available resources and fixing the exact sequence and times at which each job must be performed. Thus, scheduling is a pivotal decision activity at the operational level because it specifies how the planned production must actually be carried out over time. This definition is clearly too vague to discuss properly the application of agent-based approaches to support scheduling in manufacturing and, in particular, to analyze their possible benefits and drawbacks. Consequently, this section gives a brief overview of the most relevant characteristics of scheduling problems in manufacturing in order to provide interested readers with the essential concepts and terminology.

A widespread community of researchers has studied scheduling problems over the last 50 years, and many efforts to find efficient approaches to tackle them will certainly be attempted in the future. The reasons for such a broad interest can be found in several aspects of this class of decision problems:

• Scheduling problems have a high impact on actual production systems; efficient or smart solutions can drastically improve system performances, thus providing a valuable competitive edge.

• The variety of different scheduling models that can be associated with production systems is quite extensive; no comprehensive model can be universally adopted and be worthwhile of investigation. Scheduling situations in general, and in manufacturing in particular, differ for the objectives that should be optimized and the constraints that must be satisfied. A telling example of this...