At AllChiefs’ Sustainable Logistics practice, we recognise the potential of AI to drive improvements in the logistics sector, also in reducing emissions. But the important question is, how? Are there any cases of impact already? Let’s delve into some examples.

Introduction

Artificial Intelligence (AI) has rapidly moved from the state of futuristic speculation to a part of our daily lives. The notable advancements in AI technology over the past few years have unlocked innovations and applications we could have hardly imagined before. From enhancing healthcare diagnostics, revolutionizing customer experiences in retail to the most famous example from recent years, ChatGPT: AI’s power is evident across various sectors.

Definition of AI

To understand AI’s impact on Sustainable Logistics, we must first define AI. Rooted in computer science, AI focuses on creating systems capable of performing tasks that typically require human intelligence.

Application of AI in Logistics

Within the field of AI, several subfields exist and play critical roles in their applicability to Sustainable Logistics, with some being more relevant than others. A few commonly cited subfields include:

1. Machine Learning (ML).

This subfield involves training algorithms on large datasets to recognize patterns and make predictions. In sustainable logistics, ML can for example:

• Optimize route planning by predicting traffic patterns and identifying the most fuel-efficient paths. UPS, for example, has developed and applied their tool, ORION (On-Road Integrated Optimisation and Navigation), using machine learning algorithms. ORION optimizes delivery routes in real-time by considering factors like delivery locations, traffic patterns and package weights to minimize fuel consumption and carbon emissions of their vehicles[1].

• Improve inventory management and personalized offers by accurately forecasting demand, thus reducing waste and overproduction. H&M’s Head of Responsible AI & Data, Linda Leopold, explained that “customers can have their body 3D scanned in-store, which then generates a digital avatar, enabling customers to try different denim colors and styles virtually. Machine learning then converts the body scan into a paper pattern and measurement list. The jeans are then produced and can be picked up in-store, or delivered, a few weeks later. This is an example of personalized, on-demand manufacturing. A solution that not only solves the problem of size and fit for the customer but also leads to fewer returns and decreased CO₂e emissions.”[2]

2. Robotics

While Machine Learning (ML) enhances decision-making processes, robotics focuses on the physical execution of tasks – designing and creating intelligent machines capable of performing tasks autonomously. Some of us might have seen movies depicting robots taking over humanity. While we aren’t there yet, recent advancements by Boston Dynamics with their Atlas robots signal a significant turning point in robotics applications. See here an example of their robots performing simple and even complex human tasks in a warehouse[3].

The application of robotics offers many advantages, including automating warehouse operations to achieve faster, more efficient, and less energy-intensive processes. Amazon, for instance, has implemented robotics into their operations as they increasingly seek to automate their warehouse activities with employees fearing displacement[4]. In October 2023, Amazon started a pilot with Digit, “a two-legged robot that can grasp and elevator items”. These robots mostly execute repetitive tasks giving human employees more time to give better service to their customers.

In the last few years, you might also have seen companies deploying autonomous vehicles and delivery robots to reduce reliance on traditional, fuel-consuming transportation methods. A standout in this field, is the Swedish company Einride, a freight technology company providing digital, electric and autonomous shipping[5]. In 2019, Einride collaborated with DB Schenker in a pilot to become the first company in the world to deploy an autonomous, electric freight vehicle on a public road[6]. In January this year MARS Inc. partnered with Einride, as part of its efforts to reduce carbon emissions, to bring a fleet of 300 electric and digitally optimized, heavy-duty trucks to its network across Europe by 2030. Part of this partnership is to start an autonomous pilot in 2025[7].

The reason companies are more and more considering the transition to autonomous vehicles within their operations, is because these vehicles hold significant potential for reducing emissions due to their enhanced operational efficiency compared to traditional vehicles. Conventional vehicles, driven by humans, often consume more fuel in situations involving frequent braking and acceleration, which is common in regular traffic[8].

This driving behavior can lead to greater emissions whereas advanced computers controlling autonomous vehicles allow for a much smoother driving experience – with speed and acceleration under control and thus a more efficient fuel use – making them a more environmentally friendly alternative in terms of electricity consumed and overall air pollution generated[9].

3. AI Planning and Optimization.

Another subfield of AI which emerges as an effective way to significantly improve (logistical) operations is when using AI for planning and optimization purposes. This field focuses on creating strategies or sequences to achieve specific goals, using advanced algorithms for solving complex optimization problems. For instance, it can optimise fleet operations and maintenance schedules, thereby minimising downtime and enhancing overall efficiency.

However, sometimes optimising operations for general efficiency purposes, such as speeding up deliveries or maximising cargo loads, can inadvertently lead to higher emissions. In these cases, AI can play a crucial role by incorporating environmental impacts into decision-making. Rather than solely focusing on cost or speed, AI can help logistics managers consider the emissions produced by each option. This involves tracking and documenting emissions (emission reporting), analyzing the data to understand the sources and levels of emissions (emission analysis), and implementing strategies to reduce emissions.

What’s in it for you?

It is inspiring and promising to see that several companies have successfully integrated AI into their logistics operations, showcasing its potential. As AI continues to evolve, its role in logistics will expand and present new opportunities for innovation and the adoption of sustainable practices.

At AllChiefs’ Sustainable Logistics practice, we are keenly aware of these developments and keep a close eye on them. From our experience, we have seen a large willingness among companies to achieve their sustainability goals, but the question of how AI can support decarbonization is still hardly considered.

We are curious to hear your perspective: where do you think AI can be of value in the decarbonization of logistics? Check our LinkedIn and let us know!