Access to more products and affordable services through home delivery is convenient. But this convenience requires logistical coordination and higher costs that companies strategically try to absorb rather than pass on to consumers.

New, interdisciplinary research from Ohio State may help companies better understand their logistics costs and ultimately reduce their last mile delivery expenses.

Department of Marketing and Logistics faculty Anníbal Sodero, Vince Castillo and Walter Zinn, and Professor of Integrated Systems Engineering Marc Posner collaborated on the private and crowdsourced fleet planning capacity study.

Sodero presented the research at the prestigious Consortium on Operational Excellence in Retailing (COER) conference held in June at Harvard Business School.

The research and the solution

The law of supply and demand states that as demand goes up, supply goes down and vice versa. What many don’t realize is retailers face a last mile pricing dilemma when delivering products to consumers. They struggle with a two-sided uncertainty: uncertainty in demand of deliveries and uncertainty in the supply of available fleets and drivers.

“When you think about the total logistics costs ― to bring a product from the manufacturer to the shelf or to your home ― and you compound all of the logistics costs, a large component is the final mile of delivery, which is 30% of the total logistics cost,” Sodero said. “This is a huge issue for retailers, delivery companies and platforms to figure out how to lower those costs and still maintain a high service level when delivering products to the consumers.”

Sodero and his team researched these last mile delivery costs and options available to mitigate them through investment in a private fleet, crowdsourcing deliveries, or a combination of both.

During high- and low-demand times two-sided uncertainty and surge pricing comes into play. With food delivery companies for example, delivery demand is typically higher on Mondays and Fridays and at the beginning and end of the month, and it’s lower in the middle of the week and month. The labor, or supply side, of driver availability due to random things as weather, events or street closures, can become unreliable causing surge pricing for drivers.

“Investments in a fixed capacity like private capacity may be very costly because you may have a lot of drivers and a lot of vehicles that you end up nopt using,” Sodero said. “On the other hand, if you need to use crowdsourced drivers it may be very expensive because of surge pricing.”

Sodero and his colleagues created a mathematical solution that large and small companies, courier services and platforms can use to strategically plan and improve their service level while lowering costs.

“We developed a model that considers investments a company is going to make to calculate the optimal number of deliveries they can make using their own private fleet,” Sodero said. “The model incorporates the uncertainty in supply, costs in hiring drivers and the costs of not being able to deliver product. The model helps companies estimate how many vehicles to buy and how many crowdsourced drivers to hire.”

The researchers tested the model with companies in Brazil and the United States. The results were surprising.

Companies typically think about the lowest cost, Sodero said. If crowdsourcing is cheaper, they lean toward crowdsourced delivery; if a private fleet is cheaper, they invest in a private fleet. And increasing uncertainty doesn’t always mean investing more in a private fleet.

“Typically, what you would expect to see is the more uncertainty a company faces in demand and supply, the more investments they are going to make in private delivery to buffer against that uncertainty,” he said. “But we found that’s not generally the case. Finding a mix between crowdsourced delivery and private fleet is generally the best solution to deal with the uncertainty.”

Sodero presented the findings to COER, a group focused on advancing retail operations through cutting-edge research from both an academic and business perspective, at its annual conference, hosted by Harvard Business School and the Wharton School of the University of Pennsylvania.

“This is the most important conference I do every year,” Sodero said. “It gets me in front of practitioners ― academics and retail managers ― to address the problems companies are having from a retail and delivery perspective. These managers are the ones who are looking for and watching out for visionary research they can apply to their work.”

Sodero’s presentation of the Ohio State research was among other sessions featuring leading retailers such as Amazon, Wayfair and LinkedIn.

“The research addresses concerns the various companies had, provides a manageable solution to the cost issues they are facing, and provides contextual ways the model can be applied to different environments and different sized companies,” he said.