Anyone can perform an analysis manually. However, as the complexity of the analysis rises, so does the need to employ a computer-based process improvement tool. While spreadsheets can perform many complex calculations and help determine the operational status of most systems, their use of average numbers to represent arrivals, activity times, and resource unavailability is like using a spoon to dig a canal. Simulation, provides the equipment for complex projects.

**Accurate Depiction of Reality**

Using simulation, you can include randomness through properly identified probability distributions taken directly from study data. For example, while the time needed to perform an assembly may average 10 minutes, special orders take as many as 45 minutes to complete. Simulation allows interdependence through arrival and service events and tracks them individually. For example, while order arrivals may place items in two locations, the worker can handle only one item at a time—spreadsheet calculations assume the operator to be available simultaneously at both locations.

**Advanced Optimization Techniques**

Optimization techniques such as *linear*, *goal*, and *dynamic programming* are valuable when you want to maximize or minimize a single element (e.g., cost, utilization, revenue, or wait time). Unfortunately, these techniques limit you to only one element, often at the expense of secondary goals, and do not allow the randomness of input data (requiring you to use average process times and arrival rates)—this produces misleading results. Simulation optimization allows you to examine *multiple elements *simultaneously and track system performance with respect to activity time, arrival and exit rates, costs and revenues, and system utilization. Optimizing multiple elements provides you with the information you need to make accurate decisions and to apply more effective solutions to the entire operation.

The ProcessModel optimization module is a built-in capability that allows you to perform optimization on simulation models for Anyone can perform an analysis manually. However, as the complexity of the analysis rises, so does the need to employ a computer-based tool. The optimization module accepts parameters over which you have control and could change (e.g., the number of operators and priorities of events), and allows you to define objective functions to minimize or maximize specific model elements (through weighting factors assigned to each element). Once you identify and define these items, the optimization module performs a series of tests through multiple scenarios to seek the optimal solution. The output data details the optimized result and reports on key factors in both text and graphic forms.

**Insightful System Evaluations**

Simulation tracks events as they occur and gathers all time-related data for reporting purposes. The information available about system operations is more complete with simulation than with other techniques. With static analysis techniques such as *queuing theory* and spreadsheets, you know the average wait time and number of items in a queue but there is no way to further examine the data. With simulation, you know the wait time, number of items, minimum and maximum values, confidence interval, data distribution, and the time plot of values. It is more valuable to know that the number of items in a queue exceeds 10 only 5% of the time than to know that 2 is the average number waiting.

Static analysis techniques allow you to use only average parameters. Such limitations can mislead you with estimates that suggest an over- or under-capacity situation. For example, spreadsheets assume that production orders move unconstrained when, in fact, an operator must facilitate the move. This can yield an inaccurate capacity estimate.

**Scheduling Capabilities**

Simulation allows you to experiment with a system and see how it behaves with particular configurations of inputs, resource arrangements, routing flow rules, downtimes, and shift schedules. With the basic model elements in place, you can use simulation to test alternative production schedules through multiple scenarios and to perform many other scheduling functions.