The famous “last mile” problem explained

Parcel delivery is still inefficient and online commerce is increasing the number of parcels exponentially

As digital and structured as our modern world may be, parcel delivery is still surprisingly inefficient. In the following, we provide an overview of the reasons.

In the delivery industry, there is an age-old problem for which there is still no really efficient solution: the famous “last mile”. This refers to the very last part of the delivery process of a product, i.e. the transport section from the last distribution center to the end customer. In the next few paragraphs, we will explain why these last few meters are the reason why delivery processes are complicated and expensive. At the end of the article, we present what mobile autonomous robotics can contribute to solving the problem and also give a few reasons why all this is not quite so simple after all.

Classically little transport costs in retail

In the retail sector, it was long the case that entrepreneurs with business premises in (inner) urban areas ordered goods, usually in large quantities, from selected partners. These were delivered by the producer via various wholesalers and distribution centers of transport means with large capacities (i.e. ship, plane, railroad, truck), presented in the store, and purchased by the end customers. The delivery was organized relatively efficiently, they were Business2Business deals, the delivery people had a clear contact person (the retail entrepreneur) and a clear destination (the retailer’s business space or warehouse). Due to the typically high quantity of orders, the transportation costs per item were very low. Transportation costs are made up of components such as driver labor time, packaging, and costs for transportation equipment and fuel.

Online retail leads to exponential growth in parcel volumes

This cost structure changed abruptly due to the growing share of online retail and the resulting exponential increase in the number of private parcel deliveries. While customers value brick-and-mortar retail because of personal advice and the opportunity to try on clothes, they still often buy online. The reasons for this are the larger selection and specialization possibilities of the offer, the often lower price, the possibility of shopping without time limits and the convenience of being able to do it from home. However, the large selection offered to consumers increases the pressure on delivery: long delivery times and delivery for a fee are not accepted by customers. E-commerce thrives on satisfied customers and their positive reviews; however, delivery delays and problems lead to negative feedback, which can also ruin an online store.

For online stores and their delivery partners, the following therefore applies: on the one hand, delivery must function smoothly; on the other hand, it is precisely the last transport stage before the customer receives the goods that is the most complicated, inefficient and expensive. Depending on the calculation, the costs for the “last mile” account for between 28% and 58% of the total transport costs. But why is this so?

The small number of parcels delivered per customer (in most cases only 1-2) make for a very high time and cost per piece, these transport costs are also hardly scalable and sometimes disproportionate to the price of the goods for the retailer. This is especially true when possible returns are taken into account.

The “Last Mile” Problem

As a general rule, last-mile delivery processes are very difficult to plan. On the one hand, parcel volumes are subject to strong seasonal fluctuations; before Christmas, after shopping days such as Black Friday, or during the current COVID-19 lockdowns, they are extremely high, while in summer, when many people are on vacation, they are extremely low. The delivery companies cannot orient their infrastructure to the extremes but have to buy in external services for peak times.

On the other hand, delays in the delivery process that can hardly be calculated are to be expected, especially in inner-city areas, due to high traffic congestion, limited parking space, or incomprehensible addressing. The fact that there are other customers to be delivered to every day also makes it difficult to plan an ideal standardized route.  And finally, there is also a lot of potential for unplannability with the customers themselves. Either customers are not present at the delivery location, the delivery location was unclearly addressed or changed at the last minute, the customers have a very specific time window or special conditions (such as “please do not ring before 10:00”) at which their parcels should be delivered. Delivery boxes, such as those often found in apartment buildings, make this situation somewhat easier. The practice of many delivery companies of simply leaving parcels on the doorstep or with neighbors is questionable in terms of liability. If instead of the package, the customer only finds notification of unsuccessful delivery and a pick-up location, this means additional work for the customer, which may be reflected in a bad rating. All that was carried out for the delivery, of course, also applies to the return of parcels.

Conclusion on the “Last Mile” problem

At a time when so much is being digitized and automated, the delivery of parcels still requires an astonishing amount of work and time. Side effects of these labor-intensive and resource-intensive delivery methods include increased traffic congestion from delivery trucks and the associated exhaust and noise pollution, especially in urban areas. The working conditions of parcel delivery staff, who are subject to high time pressure, are also often criticized. So it’s no wonder that technological solutions are being feverishly sought.

Author: Lena Sophie Franke

ARTI is developing software for autonomous mobile robots. Robust enough for real-world scenarios, indoor and outdoor, and suitable for different industrial fields. ARTI offers modular AI-kits as well as full turnkey products. Our customers can pick precisely the solution they need, either build a complete robotic system from scratch or improve existing products by adding artificial intelligence.

Comments are closed.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More