Statistical Process Control (SPC) in a Restaurant - an universal Application

Statistical process control in restaurants? What does this subject line mean? It refers to a 65-page brochure published by Donald J. Wheeler, a world-renowned SPC expert, which describes the universal application of quality techniques and statistical process control in restaurants in a very interesting way. How was this done?

In the 1980s, a group of US quality experts visited various Japanese companies that had participated in a Total Quality Control Programme. As confirmation of their participation, they were awarded a Q-Mark flag, which the companies displayed in their entrance areas, and the employees also wore the Q-Mark logo on their clothing.

One evening, a group of quality experts went to a restaurant. The waitress who took their order had the 'Q-Mark logo' on her 'work clothes'. It turned out that the lady was a member of the club's internal 'Quality Circle'. She then presented a summary of a project that had been carried out in the restaurant over the past year. The aim of the project was to achieve cost savings. In addition, the participating waitresses were to improve their relationships with each other through their work on the project. The project was entitled 'Let's reduce the loss of beer and sake' and was selected after a brainstorming session within the group. Problems that led to losses included the time taken to fill glasses, correctly taking orders for drinks, etc. The project was then scheduled using a Gantt chart, starting with the selection of topics, followed by cause analysis and finally a review of the improvements with a final report. For fine-tuning the schedule, the Gantt chart was then converted into a PERT network, in which the individual process steps in the implementation were assigned daily targets.

Afterwards, the differences between the stock levels and the beverages sold were determined monthly using a Pareto chart. Sake had the most losses, beer was in second place and the combination of both was number three. The focus was placed on beer and sake. The corresponding volumes were then graphically represented in a diagram and converted into financial data (money). The causes of the losses were then determined in various cause-and-effect diagrams (service, work instructions, 'other'). The results identified, for example, 'too many tasks' for the waiters, but also things like "daydreaming" and 'lack of motivation'.  Different locations where orders/bills were filed or the incorrect operation of the sake warming machine were also identified. Corrective measures were then derived from these cause-and-effect diagrams, also known as Ishikawa diagrams. For example, a fixed place was designated for storing invoices, and training for new employees was provided by the supervisor. The success of the corrective measures was then tracked using tables and statistical process control. The result was impressive. Beer losses were reduced by 67%, and sake losses by as much as 97%.

But there were also intangible effects. Communication between waiters improved and there were fewer complaints about incorrect orders. As one of several follow-up measures, the differences between the warehouse list and the sales list were determined every fifth day of the month in order to continue monitoring losses. The entire project was carried out according to Deming's cycle: Plan-Do-Study-Act (PDSA).

Conclusion: Quality techniques, such as Statistical Process Control (SPC), are also well suited for 'beginners' and can be used (almost) universally. And that also applies to a restaurant. The brochure is available under ISBN 978-0-945320-30-2.

At the ECA Live Online Traning Trending of Process Data for OPV/CPV from 15-17 October 2025, you can learn about possible applications of SPC in the context of ongoing/continued process verification.