Berg, Douglas L.; Nagel, Robert J. (1990, ASQC) General Motors Corporation, Ypsilanti, MI
Total Quality Control requires the deployment of functions and resources throughout an organization to deliver a quality product or service to the customer. While the focus is often on the external, or final, customer, there are also internal customers, those persons or process elements that affect the goods or services produced or delivered. Each step involved in meeting the final customer's wants and needs including indirect support functions can be viewed as a "process" or "service" point where material or other inputs arrive, are handled, and are sent oil.
In many cases an aspect of the customer's perception of the quality of a product or service is the time spent in the system or waiting. This would include not only the actual processing or service time, which is usually quite short in proportion, but also time spent queued up prior to service. This is readily apparent in situations involving perishable goods or common everyday experiences such as personal banking, food service, and retail shopping. From the standpoint of the organization providing the good or service this time spent in the system can affect customer response, speed to market, throughput, in process and final inventory; staffing, and capitalization of equipment and facilities.
A body of mathematical, scientific management knowledge known as queuing theory, shows that the time spent in the system or, equivalently, the number of units in the system can be expressed as a function of the characteristics of the sequence of units arriving for processing or service and the process or service itself. Average rates are certainly important, but more significantly, the time in the system is also driven by the variability in the processing or service time. Pollaczek's equation is an example of such an expression and it resembles a familiar "smaller is better" type of quadratic loss function.
Through simulations, this paper explores the implications for improved customer satisfaction or manufacturing throughput resulting from continuous improvement in processing or service time. Reduction in variation directly affects the time spent in the system by not only making throughput more predictable, but also by reducing the expected .waiting time or the expected number of units or customers in the system. This has immediate effects not only on customer satisfaction and perceived quality but also on improved organizational response and potential reduction in operational costs and investment. Reduction in process variation, like reduction in product characteristic variation, impacts several "bottom line" factors and represents an important competitive strategy.
The simulations are based on mathematical and statistical descriptions of simple systems with single processing or servicing units. As with all models, the exact results are dependent on the assumptions behind the model as well as the distributions employed and their parameters. However, conclusions can be drawn about general concepts and results that are applicable to more complex systems, some of which would be very difficult to model. The effects on overall throughput, system responsiveness, backlogs and waiting due to common variance increasing activities, such as expediting, are demonstrated with the simulations.