2.2: Elementary Modeling Constructs

This section presents the model building perspective taken in this text. Basic modeling constructs are presented.

The operation of many systems can be effectively described by the sequence of steps taken to transform the inputs to the outputs as shown in a value stream map. This will be our perspective for model building. A sequence of steps specified in a model is called a process. A model consists of one or more such processes.

The modeling construct used to represent the part, customer, information, etc. that is transformed from an input to an output by the sequence of processing steps is an entity. Each individual entity is tracked as it moves through the processing steps. Processing can be unique for each entity with respect to such things as processing times and route through the steps. Thus, it is essential to be able to differentiate among the entities. This is accomplished by associating a unique set of quantities, called attributes, with each entity. Typical attributes include time of arrival to the system and type of part, customer, or other information.

Note that a value stream map shows in general how parts flow through a system. This might be viewed as a “macro” or big picture view of how a system operates. One characteristic of beyond lean is the use of models that give a more detailed or “micro” representation of a system. This helps in gaining of understanding of how a future state will actually operate and aids in ensuring a successful transformation.

Certain components of a system are required in performing a processing step on an entity. However, such a component may be scarce, that is of insufficient number to be available to every entity upon demand. Such a component is called a resource. Waiting or queuing by entities for a resource may be required. Typical resources may be machines or workers. Other system components, totes or WIP racks, may be scarce and modeled using resources.

A resource has at least two states, unique circumstances in which it can be. One of these is busy, for example in use operating on an entity. Another is idle, available for use. Typical model logic requires an entity to wait for the resource (or resources) required for a particular processing step to be in the idle state before beginning that step. When the processing step is begun, the resources enter the busy state. As many resource states as necessary may be defined and used in a model. Other common resource states are broken and under repair, off shift and unavailable, and in setup for a new part type.

Consider a workstation consisting of two machines. A basic modeling issue is: Should each machine be modeled using a distinct resource? Often, it does not matter which of the two machines is used to process an entity and operating information such as utilization is not required for each machine. In such a case, it is simpler to use a single resource to model the two machines. However, an additional concept: number of units, is necessary to model two (or mores) machines with one resource. There is one unit of the resource for each machine, two in this example.

State variables, and their values, describe the conditions in a system at any particular time. State variables must be included in a simulation model and typically include the number of units of each resource in each possible resource state as well as the number of entities waiting for idle units of each resource, the number of entities that have completed processing, and inventory levels.

Time is a significant part of simulation models of systems. For example, when in simulated time each process step begins and ends for each entity processed by that step must be determined. Sequencing what happens to each entity correctly in time, and thus correctly with respect to what happens to all other entities, must be accomplished.

In summary, modeling the behavior of a system over time requires specifying how the entities use the resources to perform the steps of a process as well as how the values of the entity attributes and of the state variables, including the state of the units of each resource, change.