IBM ILOG Scheduler User's Manual > Advanced Concepts > Advanced Features and Concepts > Transition Costs

Between each pair of consecutive activities on a unary resource, some cost may be incurred to switch the resource from processing one activity to processing the next. These costs may be related to modifications to the resource, such as adjusting or purging a machine. The modifications may require manpower, material, and energy.

In Scheduler, transition cost is defined as the cost between two immediately successive activities scheduled on a unary resource. In addition, Scheduler lets you define a setup cost for the activity that starts the usage of the resource and a teardown cost for the activity that ends the usage of the resource.

Two types of transition times and costs can be modeled in Scheduler: constant and variable.

Constant transition costs only depend upon the precedence relationship between two activities. In that case, as shown in Adding Transition Times, Scheduler associates a transition type with each activity. The accessors are the functions IloActivity::getTransitionType and IloActivity::setTransitionType. These transition types allow you to define constant transition times and costs that are represented by the class IloTransitionParam. This class consists of a table of integers indexed by the transition type of activities.

The evaluation of variable transition times and costs depends upon current knowledge about the other variables and constraints involved. The macros ILOTRANSITIONTIMEOBJECT and ILOTRANSITIONCOSTOBJECT allow you to define variable transition times and costs in Scheduler. More precisely, these macros allow you to define and implement the extraction of an IloTransitionTimeObject or IloTransitionCostObject into your own IlcTransitionTimeObject or IlcTransitionCostObject, respectively. In the case of variable transition costs, the Ilc version of the transition cost object must define its minimal and maximal value given the current knowledge about the sequence.