The Trade-Off: Service Vs. Costs

To maintain or grow market share, retailers must be active in e-commerce and home delivery. This marks a considerable shift from traditional store operations, especially when it comes to the process and cost of last mile delivery. All e-logistics companies face losses in their first years of operation, as we saw with Amazon and Zalando. At the same time, customers are expecting better service: We want to book our own time slot, we want to know the expected time of delivery and want to be able to return goods easily and for free. But if everybody in our neighborhood picks their own desired time slot, the delivery van must return multiple times to our area, which implies additional costs. The more time slots a customer can select from during the day, the higher the service but also its delivery costs. How can (grocery) retailers manage this?

Enter Dynamic Time Slot Booking

Things would be simpler and more efficient if customers were not able to pick a time slot and got an estimated time of arrival instead. But most of us don’t want to be home all day waiting for groceries to arrive or goods to be installed. So, to accommodate this need, retailers now offer delivery time slots. Possible time slots can differ per area to create efficiency. Open and available time slots are typically calculated in a split second on an app or website, and most companies are using simple capacity rules to do this calculation. For example, they base the availability of time slots on a maximum number of stops per hour.

However, this is often not enough to mitigate the high costs. A time slot should be made more attractive for a customer to pick if there are orders booked in the vicinity, or expected orders to be delivered close by. This would make it easier to fit orders together in a single route, making the delivery more efficient and less costly. If booked orders are clustered in a certain region, there might also be room for additional bookings. When simple capacity rules are not enough to do this type of clustering, you can use dynamic time slotting.

ORTEC - Dynamic Time Slot Booking Grocery Delivery

Goos Kant – Professor Logistics Optimization

"A time slot should be made more attractive for a customer to pick if there are orders booked in the vicinity, or expected orders to be delivered close by."

How Dynamic Time Slot Booking Improves Service and Efficiency in E-Logistics

Dynamic time slot booking uses optimization software to continuously create optimal routes based on booked (and forecasted) orders. For every new booking request, the optimizer checks all insertion options in these routes and calculates the additional costs for each feasible time slot. The solution should compute all these possibilities within a split second, so that concurrent customers can book delivery time slots without a problem.

From Trade-Off to Benefit: Improve Costs and Service Simultaneously

Research shows that booking time slots dynamically can lower costs considerably. In one particular case, a retailer combined the possibility to book a dynamic time slot with “green labels” that displayed attractive time slots more prominently for customers. These available time slots had a low insertion cost but were also more environmentally sustainable to execute. The result was a 9-13% decrease in delivery costs per order in urban and rural areas respectively.

Our own simulations also show that 5-10% more orders can be booked using the same capacity, when booking a time slot dynamically. Because available capacity is used more efficiently, 30% more time slots can be offered during requests. For instance, Waitrose, a leading British retailer, saw a significant increase in the number of deliveries without increasing the van capacity. This enhances the customer’s service perception and increases the company’s Net Promoter Score. In other words, it enhances both service and efficiency.

Dynamic Time Slot Booking Offers More Opportunities

If you’re curious about dynamic time slot booking and you’re wondering how to make it attractive to customers, consider running an experiment. For instance, you can also offer only low budget variants, where only the cheapest (few) options are shown to the customer. By testing different strategies to influence the consumer, you can gain insights into expected service and efficiency. Using this method will also give you insights continuously during the booking process, so you can better understand how to allocate resources and capacity. You may discover more effective ways to process booking requests or even make gains in other areas, such as picking orders earlier in the warehouse. All of these benefits can lead to significant savings in last mile e-logistics, where every second counts.

About the author

Goos Kant (1967) is full professor of Logistic Optimization at Tilburg University. He is involved in the master program of Business Analytics and Operations Research, as well as in the master program Data Science & Entrepreneurship. He is the project leader of a large R&D project on horizontal collaboration, speaker at conferences and teaching in executive education programs. Goos is also managing partner at ORTEC, and involved as optimization expert at C-level with their customers. He holds both an MSc and a PhD in Computer Science.

CONNECT WITH GOOS
ORTEC - Goos Kant professor of Logistic Optimization at Tilburg University & managing partner at ORTEC

Don't miss out on the next insights

Sign up to our mailing list and be the first to receive our newest insights and digital magazine in your mailbox