In the bustling city of Rio de Janeiro, opening and operating a series of barbershops comes with a unique set of challenges, especially when operating on a tight budget for renovation and rental expenses. In this article, I am sharing the story of one entrepreneur, the owner of multiple barbershops in the city, who leveraged SCDA’s analytics competence for drafting a service facility allocation and expansion strategy for revenue and profit maximization. The findings from this project are relevant to all owners and operators of service facilities (beauty salons, barbershops, repair shops etc.), points of sale (gadget stores, groceries, etc.), and e.g. also restaurants, hotels, AirBnB rentals, and coffee shops.

Analytics as a guiding tool for service facility allocation

Faced with the task of opening a series of barbershops for a new branch in Rio de Janeiro, the entrepreneur turned to prescriptive analytics as a guiding tool. Together with SCDA a mathematical model was developed. Deployment took place in the form of a web-app for long-term user access. This allowed the user to re-run the underlying optimization model for updated data and scenarios, by adjusting parameters. This also allows the user to apply the underlying model and analytics to other locations, e.g. another city or country.

The mathematical optimization model for service facility allocation considered the following aspects:

• Regional market potential: Different regions in Rio de Janeiro have different market potential for the services offered by the barbershop operator, which is a product of potential consumers living, working, or being accommodated (tourists) in the respective area, and their projected average spending budget. In some areas of the city, consumers have significantly higher spending
  budgets compared to other areas of the city.
• Proximity-based consumer purchasing likelihood: As the barbershop operator positions its business as a walk-in barbershop business, each potential customer is more likely to visit the shop if the shop is nearby.
• Limited renovation budget: The maximum amount of barbershops depends, partially, on the maximum renovation investment budget available to the owner.
• Limited rental expense budget: Another constraint is the owner’s budget for monthly rent payments for renting the respective store.
• Limited daily capacity: A barbershop has a certain capacity limit to how many clients can be serviced, in this case mainly based on the size of the shop.

The barbershop chain owner already had a list of potential locations that were available to rent. Potential sites had been identified by real-estate agents. In addition, a market and service demand forecast was developed. This was done by a third party. The forecast mapped potential consumer density and average spending budget per consumer to spatial reference points, resulting a fine-meshed map-based market potential forecast. The SCDA facility allocator app accepts this demand forecast as model input, and visualizes the spatially distributed market potential on a map, using a heat map. See below.

Above map visualizes the market potential in the form of a heat map, with markers for all potential barbershop locations identified by the barbershop operator’s real estate agent. The potential sites represented by these markers are provided as another input table (.csv file), considering capacity per site, rental expense and investment need per site, and spatial location coordinates.

Drafting expansion by analyzing locations for different budgets

While it would be extremely difficult to forecast actually realized revenues and net profits precisely, the mathematical model is great for receiving recommendations for optimal locations for a given investment and rental expense budget, assuming a defined spatial demand distribution. This is how the service facility allocator app was used. Below figure shows the optimal locations for different budget levels, simulating an expansion strategy. The initial budget for location setup was 200,000 BRL, and the budget for monthly rental expenses was originally 20,000 BRL.

Based on the results from the sensitivity analysis above, three (3) locations were selected for initial opening (phase #1), and another six (6) locations were selected for phase #2 and phase #3 expansion, three per phase respectively. This covered an expansion plan from an initial 200,000 BRL location investment and 20,000 BRL monthly rent expenses, up to 1,000,000 BRL location investment and 100,000 BRL monthly rent expenses.

This is a perfect example of how the allocator app should be used: The app does not tell the user what the optimal spending budget for rental expenses and facility renovation is. Instead, it assumes that the full budget for rental expenses and renovation expenses may be used. It then tries to maximize revenue by selecting facility locations that reach the largest possible group of consumers with the largest possible consumption budgets.

In order to develop an expansion plan, the tool should be used to run a sensitivity analysis – such as the one above. In this way the app guides the user towards identifying an optimal expansion plan, starting with the most promising and likely most profitable locations first, and after that expanding with additional but likely less profitable locations. Doing this step-by-step, with active management and monitoring, will give owners and investors the best means to maximizing returns on investment.

Concluding remarks and related content

If you are facing similar facility allocation problems, related to sales stores, service centers, repair shops, restaurants, or similar, you can try downloading the facility locator app on the SCDA website: Service facility allocator. Below video demonstrates the use of the app.

The app is currently downloadable as a deployable Python project. You can deploy it to a Shiny server – where it can run as a web-app useable by authorized users. You can also obtain the app in already deployed state, so that you do not have to handle the deployment. For that, you need to contact the SCDA team here: Contact

Linnart Felkl

Data scientist focusing on simulation, optimization and modeling in R, SQL, VBA and Python