At 12.5 hours of work a day, nine mobile shops face a simple question with a complicated answer: when should they leave one stop and drive to the next? Researchers at the University of Texas at Austin have built a model to help retailers make that call, balancing customer demand, travel time and the freshness of their goods.
The question matters as shops on wheels move beyond the familiar ice cream truck. Some mobile retailers sell baked goods and fresh meals, and both piloted and self-driving stores are being developed. BPM Analytics projects that worldwide sales from retail on wheels will reach $30 billion by 2030. For small businesses, a mobile store may also offer a way to reach customers without the high costs of opening and running a permanent shop.
Junyu Cao, an assistant professor at the McCombs School of Business at the University of Texas at Austin, worked with McCombs doctoral student Zhuolun Dong and Wei Qi of Tsinghua University on the research. Their mathematical model weighs whether a vehicle should “perch,” or stay put and sell, or “migrate” to a new location. To build it, they drew on real-world data from Neolix, a Beijing-based company whose autonomous vehicles deliver meals to workers in office parks and industrial areas.
The model considers how quickly customers at a stop make their purchases and demand fades, how fast food loses freshness, and how much ground each vehicle can cover in a day. The researchers found that timing matters. Early in the day, when products are freshest, stores should move often and visit more locations. Later, as goods grow less fresh, they should stay longer at each stop and serve the demand still available there.
The best plan also depends on the crowd and the pattern of sales. In densely populated areas, a store can focus on busy places instead of trying to cover a wide map. When customers arrive steadily rather than in a sharp rush, staying in one place can be better than spending time on the road. Adding vehicles can pay off during busy periods, when each shop can serve a separate high-demand area. When sales are slower, however, extra vehicles may compete for the same customers, reducing the benefit of expanding the fleet.
Cao says the findings could help operators plan routes and make traditional food trucks more flexible. Many trucks tend to remain in one place; moving more strategically could help them reach fresh groups of customers. The model may be especially useful for small retailers that cannot easily afford a permanent storefront. Cao describes mobile stores as a lower-cost option that lets businesses use movement to reach buyers.
The study, “Retail on Wheels: A Time-Sensitive Traveling Salesman Problem,” by Dong and co-authors, is forthcoming in the journal Management Science. Its central lesson is not simply to keep moving: a mobile shop earns more by matching its route to the hour, the crowd and how quickly its products lose freshness.