Mercator Endowed Chair of Demand Management & Sustainable Transport

Making Use Of Dynamic Time Slot Management

Online retailers face significant challenges due to the corona crisis.

With the Corona crisis, the popularity of ordering food products online has risen significantly, which poses considerable challenges for online retailers. New approaches in managing delivery or pick-up windows could help to serve more customers and prioritise certain customer groups (e.g. high-risk groups).

The British e-commerce grocery retailer Ocado reported in an email to its customers in late March that virtually every customer, who has ever placed an order in the past, tries to place an order on a weekly basis now. At the same time, the average order quantity is doubling, and the queue on the company's website sometimes numbers several hundred thousand customers.

This sudden surge in demand poses a significant challenge to retailers. To address this situation, Ocado, along many other grocery retailers in the UK, has not accepted new customers, allows a maximum of one delivery per week per customer, and has also prioritised customers in high-risk groups and those deemed most valuable to the company. They are being sent links to help them bypass the ordering queue. However, the customer still has to select a one-hour delivery window from a number of available options.

Recently published research by the WHU ­­­­- Otto Beisheim School of Management in collaboration with the University of Warwick shows that more flexible time slots would allow for more customers to be supplied (not only in times of crisis, but particularly under “normal” demand levels). Essentially, the idea is that a customer can select several time slots according to personal availability, e.g. 8-9 a.m. and 6-7 p.m. The day before the day of delivery, the customer is then going to be notified in which of the selected time slots the delivery will arrive. In return, an incentive is granted e.g. a reduction in delivery charges. It is important that the incentives granted are balanced with regard to the benefits (gained through the additional flexibility). This can be ensured by newly developed algorithms. Real-time optimisation is used to determine which windows are available at which (dynamic) prices for a given order.

Since many customers currently work from home, a lot more people should also have more flexibility with regard to delivery time windows. According to a survey by the BITD in Germany, there will also be more people working from home after Corona: more than two thirds of the employees surveyed expressed an interest in this. Therefore, using this flexibility seems to make sense and would moreover serve sustainability. The study shows that more flexibility in delivery time windows generally also leads to more efficient delivery plans. Especially in times of crisis, many customers will certainly be prepared to accept greater flexibility so that more customers can be served overall if this is precisely communicated. The greatest potential for improvement is to be expected when demand has gotten back to its regular level. At this time, route planning will benefit the most. This investment would therefore also make sense in the medium and long term.

Demand for Click & Collect is also exploding, exceeding available capacities many times over, retailers report. There, too, time slots are suggested for collection, but are usually allocated on a first-come-first-served basis. In times of extremely high demand, an approach that dynamically reserves and releases capacity for customer groups with priority could be suitable here – comparable to the approaches used by airlines. According to Professor Arne Strauss of WHU, such an approach could help in the supply of priority customer groups such as health care personnel, especially when the demand from these groups is variable and uncertain. Additionally, some of these customers may also be more limited when it comes to the availability of their time. Under these circumstances, techniques from the airline sector could be quite useful to contribute to the supply of priority groups. Such a procedure might also be helpful after the crisis, e.g. by defining priority groups via their frequency and volume of ordering as well as the segment-typical time slot selection behavior.

Thus, the crisis may be seen as a good opportunity to strengthen the online food sector by improving automated time slot management.