The rise of e-commerce in food shopping has been accompanied, for nearly 20 years, by a multiplication of delivery methods and the possibilities offered to consumers. These new methods today raise fundamental questions of relevance regarding their adoption by consumers, their profitability prospects for distributors, but also their societal, economic and ecological effects.

Among all these collection models, robotic lockers* seem to stand out to the point of producing a genuine flurry of different concepts in recent years.

Many large retail brands are experimenting with them at the local level to better understand their uses by customers and their economic profitability, without pushing the experimentation to large-scale deployment. If the idea of the relevance of robotic lockers is gaining ground, we have not yet seen a real consensus forming on an optimal mode of operation or an inflection point in the adoption of this method of withdrawal by large-scale retailers. . To find a viable business model, it is essential to think beyond the instructions themselves and to think about the complete logistics plan on a local scale.

An overall logistical vision

Robotic lockers, drive, home delivery and click&collect represent a complete and complementary range of collection and delivery methods, offered alongside traditional shopping in large and medium-sized stores. The idea is not to focus on one channel rather than another, but to offer a range of services capable of responding to as many customers' daily cases as possible.

The relevance of robotic lockers lies in its ability to offer a solution that goes beyond the geographical and temporal limits of other collection methods. Drives, home deliveries or click&collect collections involve travel to or from a commercial area as well as a prescribed time slot for collection. Offering more flexibility, deposit collection can be done at any time and potentially in pedestrian mode.

Alongside the emergence of these new modes of consumption, the hypermarket and supermarket model is losing momentum and is forced to reinvent itself. New physical and digital flows call for a transformation of these sales areas into spaces capable of addressing all collection and delivery channels. We see these areas evolving into real Micro Fulfillment Centers** to which the addition of new modes makes it possible to gain and maintain market share. The objective is then to make this multi-channel supermarket model profitable.

What levers can be used to ensure the viability of the economic model?

In the “classic” shopping model, the preparation of orders and delivery to the consumer’s home is at the customers’ expense. With the drive, home delivery and robotic lockers, it is the brand that takes care of all preparation operations and all or part of the delivery costs. To develop their business while remaining competitive, distribution brands must control this increase in operational costs. Two levers are at their disposal:

The first is the centralization and automation of order preparation operations for deposit collection, drive and home delivery. On the one hand for reasons of compactness – land is rare and expensive, on the other hand for reasons of operational productivity. Investment in a robotic picking and storage system is essential to achieve high operational performance in a reduced volume, and therefore make the activity profitable.

The second lever is the establishment of an information system capable of orchestrating all processes, from product storage to feeding the different collection and delivery methods. Software intelligence is key to improve the efficiency of each mode: responsiveness when a customer arrives in a Drive, shortest route for a home delivery round, etc.

The management of robotic instructions must provide answers to two key points. We are working in particular on predictive management of slot occupancy time, with the aim of estimating the number of free slots and improving the profitability of the system. Learning by statistical observation is a first step, and it is possible to go further by estimating, via the use of machine learning, a withdrawal time based on the content of the orders. We are also considering automated verification of returns made by customers via the instructions. This will go through a product control process which may involve scanning the barcode of the items, weighing them or even automatically recognizing objects using a vision system.

A beneficial impact in reducing CO2 emissions

Beyond the economic model, there is also the question of the environmental impact of withdrawal methods, and it must be said that robotic lockers have arguments to put forward in this area.

Let's take the most common scenario, namely shopping carried out on a weekly basis, with the desire to have a large choice of products. Home delivery is traditionally the method with the lowest CO2 footprint, due to the pooling carried out via an optimized delivery round, where the drive and “classic” shopping generate a large number of trips between supermarkets and homes.

Providing a network of robotic lockers is even more interesting than delivering to customers one by one. This is the effect of the massification of delivery points brought by this new mode. With instructions judiciously positioned to generate withdrawals in pedestrian mode or on journeys between home and work, the CO2 reduction reaches up to 40% compared to home delivery.

In fact, the contribution of robotic instructions to CO2 reduction differs depending on the density of the habitat. In an urban area, the locker will be located in a dedicated space or more likely in a space combining a mini-market and a locker: shopping is collected on foot or via a journey from home to locker using soft transport. In peri-urban areas, lockers must be positioned in transit areas between homes and workplaces to minimize detours made to collect shopping. The only exception to this table, the use of deposits in rural areas does not bring substantial gains in terms of carbon footprint, but represents a service strengthening the attractiveness of the territories.

Conclusion

As we have discussed, robotic locker presents a major opportunity to the host of pickup and delivery methods that have emerged in recent years. Although it presents solid arguments, many questions still surround it and are, for the moment, slowing down the democratization of its use and its large-scale deployment. Current experiments encourage us to rethink so-called last mile logistics models beyond the deposit alone, but in connection with its method of supply and order preparation. It also remains to clarify the ideal layout plans for collection points and MFCs, as well as the order profiles and responsiveness and flow objectives expected by customers for each location.

Once these questions have been resolved, it is a safe bet that these systems will be successful, as the benefits in terms of customer service and ecology are so important in a global logistics vision.

To find out more about SAVOYE, visit our website:  savoye.com

Notes:

*: To find out more about robotic instructions: https://www.savoye.com/ressource/consignes-robotisees/).

**: To learn more about Micro Fulfillment Center: https://www.savoye.com/solutions-2/micro-fulfillment-center/).

Read all the « Expert Opinion » articles on the SprintProject blog