Analyzing historical and experimental data can have a powerful effect on a business’s operations, but what happens if no good data exists? Most companies are reduced to guesswork and hope, but by translating personnel’s knowledge into a set of reasonable assumptions and applying the proper theoretical framework, much more insight can be gained through a simulation. This insight allows businesses to understand if their estimations of end results are consistent with their assumptions, and goes beyond estimating the most likely result to also illustrate the range of probable outcomes. Here are some examples of simulation and modeling and how they were used:
Equipment Capacity Simulation – A manufacturer had a process that was routinely not able to hit their production demands and was considering purchasing a new, and very expensive, piece of equipment. Prior to doing this, they decided to model their current equipment’s capacity. The model showed that the current piece of equipment was only operating at 60% of its theoretical capacity. Furthermore, the model showed that with some simple changes, an additional 30% of capacity could be gained. In short, they were able to double their capacity and avoid the costly equipment purchase through improved production efficiency. Since they simulated the process, they had the necessary insight and confidence to do this.
New Product Evaluator – When a client had more new product ideas than they could successfully convert into products, they needed a way to evaluate the candidates. Borrowing from the best practices, we developed an application that considered strategic business factors, but more sophisticatedly used a Monte Carlo simulation within the application to model both the most likely sales, cost, and price outcomes along with possible ranges. From this, we calculated likely profits and, more critically, a risk metric and the probability of the product failing in the marketplace. This insight reduced the cost and time associated with unsuccessful product launches.
Inventory Requirement Simulation – A manufacturer was in the process of expanding to a larger facility and was also purchasing automated equipment. Since the units they made had highly variable work content, they needed some level of work-in-process (WIP) on the floor, but wanted to keep the quantities to a minimum. Combining their historical production data with a model created from the new equipment supplier’s machine specifications allowed for the creation of a stochastic “WIP needed” model. This allowed them to design the new factory floor layout in an optimum fashion to minimize inventory and maximize flow.