ISE Magazine Volume : 49 Number: 12
By Juan Martinez
Cheaper, expanded testing allows restaurants to innovate with menus and services
Industrial and systems engineers have been applying simulation in general industry for a long time. Decades ago, spreadsheets, operations research models and computer simulation using GPSS (general purpose simulation system) were considered cutting-edge techniques. More advanced principles and simulation technologies have been developed since then, and comparing the two eras is like comparing an abacus to a calculator. Considering how costs, especially labor costs, keep creeping up in the industry at a fast pace, applying industrial engineering principles in restaurants can drive significant impact to bottom-line profits and sales processing capacity in restaurants. As evolution has taken its place, not only has this made simulation software more friendly to the users and more powerful in using modeling for analysis, it also has facilitated understanding and acceptance by nontechnical and management personnel.
How simulation benefits food service
The areas that could be simulated include customer service flow, kitchen line production, drive-through and other modes of service, dining room seating capacity, ordering capacity (eat-in, online, delivery, etc.) and other areas. Another significant area where industrial and systems engineers can apply simulation is in developing labor deployment guidelines for restaurants. Computer simulation enables you to add the dynamic (service) aspect to labor guidelines development since the system runs in real time, including delays in production and service times.
Although many may look at undertaking a labor project as an effort to reduce labor costs, the best goal in a labor initiative is to develop guides that facilitate the deployment of the right labor in the right place at the right time to drive sales, throughput, quality and an optimum customer hospitality experience. Considering that the minimum wage keeps going up and up, and the restaurant industry relies a lot on this source of labor, this is a critical way to ensure the best “unit economics” for the concept that drives maximum return-on investment for shareholders.
30 percent higher throughput
In this case, a fast-casual restaurant chain wanted to test different service systems, along with kitchen layouts and labor deployment setups, that would enable its franchisees to process more guests using the same number of employees. Fast casual restaurants generally refer to restaurants that offer customer service similar to fast-food restaurants but often with higher quality and more expensive menu items.
The service systems tested affected the inside customers as well as the drive-through guests, making testing in a real restaurant expensive, time-consuming and very limiting in the number of options that could be examined. For this particular concept, the time of the year (season) and the part of the country (region) significantly affected customer orders, so the options had to be tested for those variables as well.
The team used simulation software to create the model, which provided quantifiable differences between tested options and types of guests for service times, resource utilization and labor utilization. The simulation also was used to do sensitivity analysis to determine the impact of how well the guests would accept the new technology that was being applied. The model was used to determine the ideal number of order points needed as well as the number of new equipment pieces required to meet a targeted throughput volume, all while maintaining or improving service times.
The simulation was used to run options for different labor staffing levels ranging from two to eight people, along with different task assignments or slide deployment practices to determine the best use of the labor resource for new and existing stores. According to the simulation’s results, an alternative ordering system could achieve up to 30 percent higher throughput with the same number of employees, all while maintaining or improving service times.
Target: Operating and capital costs
Simulation was applied to test many different kitchen line layouts in order to derive one that balanced all the key metrics of the design. These metrics included labor deployment, equipment placement and cost, production and assembly times, some product quality metrics and other variables.
In this case study, the full sequence of customer service and production was tested. The simulation team measured the speed of service the guests received, including line time, order and pay, and time for the customers to receive the food, the key output metric that the team was trying to control. To meet a specific speed of service goal, the simulation team varied the design in the back production line, taking into account the location of the equipment, the type of equipment and cooking characteristics for each, the deployment responsibility for each employee and other variables.
Once the design was finished, the restaurant chain gave it a real-world test by implementing the model in a location that was due for renovation.
Simulation allows concepts to change continuously
Dynamic computer simulation is without a doubt an innovative way to test and validate food service designs. This technique has been around for some time, but it has taken a while to garner mainstream application. The benefits include:
- The ability to test more options rapidly
- The ability to test riskier options
- Less destructive and simpler testing
- The ability to continuously test on a permanent basis
- The flexibility to make changes
- The ability to add a dynamic extension to a deterministic process
The bottom line is that dynamic computer simulation provides an easy way to design and test the complexity inherent in restaurants and the different options that should be considered to grow the brand. With the use of this technique, you can develop and expeditiously test “the design of the future” today at lower risk and lower cost compared to other testing options, all while considering innumerable variables.
Published at :