Data Science & IA

Staffing optimization model for a fast food chain

The challenge

A restaurant holding company with 5 different brands, more than 150 locations throughout Chile, and more than 3,000 workers, sought to optimize the allocation of staffing and shifts at its points of sale. The company was generating shifts in a manual and discretional way, which caused multiple risks, such as inefficiencies, errors and higher costs. The scale of the business made it imperative to find a different solution to properly allocate employees in the various stores. In particular, it was established as a goal to generate an automated shift optimization model that considers multiple variables (e.g. process times, projected sales, labor legislation, among others) to determine the appropriate number of personnel per location, at all times, to ensure the defined levels of attention and service, minimizing costs, complying with labor legislation and ensuring fair and orderly shifts for employees.

With this new model, it was possible to eliminate the discretionary power of supervisors when assigning personnel and to make better use of personnel hours throughout the country.

La estrategia

Brain Food staff became involved in the operation of each of the chains studied, understanding the difficulties and particularities of the work of the people to be modeled. Thousands of measurements were taken of product processing times and of the administrative processes within each store, to obtain a significant sample for each one of them, and thus estimate how long it took on average to process each product in each of the stores studied. Taking these inputs and together with other variables (such as labor legislation and a productive sales model created by Brain Food), an automated model was generated and made available in the Brain Food suite, where the model can be run, parameters can be changed and the results obtained can be visualized.

Diagnóstico

In-depth analysis of premises operation with actual endowment utilization

Optimización

Automated model that allows defining optimal allocations per location according to context

Herramienta

Application that allowed for manual inputs and intuitive operation by decision-makers

The achievements

With the optimal staffing, improvements were achieved that gave positive results within the optimization of processes in the company. It was possible to:

  • Standardize processes for all stores
  • Significantly reduce overtime in the stores by considering a productive sales model as an input.
  • Redesign certain internal processes of the stores to improve the efficiency and simplicity of the operation.

Individual models were created for each of the stores incorporating the particularities of each one of them such as:

Differentiated opening hours

Proximity to warehouses from which supplies are replenished

Physical layout of the premises

Number of boxes

Number of production lines

Distribution of sales in the product mix