This post is also available in: Español (Spanish)
Logistics and transport are undergoing a profound change. Sixty-five per cent of industry managers (according to Forbes Insights) believe that there is almost a renaissance of the industry. The culprits? Artificial intelligence and machine learning.
Artificial intelligence in logistics
“In the logistics sector, AI can help to optimise delivery routes, improve fuel efficiency and reduce delivery times“, says McKinsey’s report ‘The promise and challenge of the age of artificial intelligence‘. According to the analysis company, the industry will be the second largest beneficiary of the AI revolution in the short term. They estimate that the potential for improving business efficiency through artificial intelligence is 89%.
Among the challenges faced by logistics companies to implement AI are the lack of a clear strategy, difficulties in finding the necessary expert talent and the company’s own organisational structure, usually organised in silos isolated from each other. But let’s go with the benefits. This is all that artificial intelligence can achieve at the controls of a delivery truck.
An efficient warehouse
In logistics, simulations play an essential role. Broadly speaking, it is the application of models to predict probable results in a given system and to correct inefficiencies. Artificial intelligence is giving new impetus to this way of anticipating problems. Let’s start with the warehouse.
H technology, an AI that Hitachi has been developing in-house since 2015, helps the Japanese company’s employees work more efficiently in its warehouses. It analyses their decisions, the likely outcomes of each of them based on multiple variables, and presents its conclusions. The company claims that its efficiency has increased by 8% in AI warehouses.
When it comes to placing, picking and moving goods in a warehouse, AI also has a lot to say. In some cases, through robots, as we will see later. But in others in collaboration with human workers. This is the case of the German Zalando. In its warehouses, an AI is in charge of optimising the routes so that the employees make the least number of movements in the shortest time possible.
The robots that deliver packages
More than the metal humanoids we imagined years ago, robots are turning out to be almost any kind of machine. They are, however, becoming more and more intelligent. In logistics, robots have started to conquer the warehouses. The cases of Amazon or Cainiao, the logistics subsidiary of Alibaba, are well known. But there are many others. The Dutch startup Fizyr, for example, has developed robots equipped with a deep learning algorithm capable of identifying, analysing, counting, collecting and handling all types of goods.
Moreover, in the future, robotics will have a lot to say outside the warehouse. From the transport of goods thanks to autonomous vehicles (such as those already being tested by UPS thanks to the TuSimple startup) to the so-called last mile logistics (such as the Scout delivery robots from Amazon).
Demand forecasting and route optimisation
Today, the great capacity of AI, the most developed, is to analyse immense data sets and use the information to predict the future. Or at least get pretty close to it. AI is increasingly better at anticipating unforeseen events and analysing likely risks. In logistics, this can be applied to both route optimisation and demand forecasting.
The German company DHL, in collaboration with IBM, is one of the pioneers in both aspects. It has been working on route optimisation software since the early 2000s. Today, its AI algorithms analyse an infinite number of variables, from the state of the traffic to the wind, in order to plan the fastest and most economical delivery routes. Another of its tools with artificial intelligence, the Global Trade Barometer, allows the company to have an adjusted prediction of the global demand for goods transport for the following two months.
In addition, as in many other industries, AI can be used to analyse and optimise internal processes, contribute to strategic decision-making, predict and minimise risks or strengthen cyber security. The intelligent logistics revolution has only just begun. As more things are connected to the internet, more data becomes available and AI training improves, the number of applications will skyrocket.