Still, the topic of how artificial intelligence will affect the road freight transport industry is an interesting one to explore. After all, the industry has jumped leaps and bounds to improve its efficiency: whether it would be by simply moving from paper to computer-based planning, or by including more complex transport management systems (TMS), there is no doubt that the old ways of orchestrating companies’ fleets have been gathering dust for in archives’ for some time now.
The integration of AI, which is defined as “the ability of a machine to display human-like capabilities such as reasoning, learning, planning, and creativity,” per the European Parliament’s definition, is becoming ever more present in our daily lives. A smartphone assistant, your translation application of choice, even the search engine of your liking adapts to our choices and decisions daily. AI-based technologies have begun appearing on our roads as well.
While they are not the true self-driving machines that some marketing slogans could falsely point out, they are capable of allowing the driver to not operate the vehicle while still requiring the full attention of a person behind the wheel at all times.
An industry-changing technology
The development and incorporation of vehicles that are fully capable of making decisions on the road are undoubtedly interesting to follow, as it would change the fundamental way that road transport interacts with other road users. The possibilities are seemingly endless, as startups are promising a future where an autonomous car would pick one up and drive them to their intended location. Current manufacturers are also looking into the future, including those that make trucks at this right moment.
“There is a revolution taking place, but at a relaxed pace. Calmly and safely, the Mercedes-Benz Future Truck 2025 rolls down the highway at 85 km/h. The tractor and trailer brake and accelerate with precision, riding in the middle of the right-hand lane in flowing traffic. This scene, however, is anything but routine,” the German carmaker Mercedes-Benz painted a picture of the future of road transport by presenting its Future Truck 2025 concept. According to the Stuttgart-based manufacturer, its futuristic concept could transform a road-driven vehicle into an office that’s on the road, allowing drivers to “perform other tasks of considerable value to their companies, including, for example, flexible scheduling of the current trip, planning of upcoming trips and bookkeeping.”
“A truck driver can completely count on the computer systems of his truck, which, thanks to its sensors and exchanging of data with its environment, safely and efficiently heads toward its destination.”
Using the driver’s working hours more efficiently is a development that every logistics firm is constantly looking at, whether it would be by the reduction of empty kilometer driving or connecting loads to the truck quicker than usual. However, the fact that a truck driver theoretically could be able to plan his working trips and schedule would fundamentally change the way that road transport is organized. Largely benefitting small, one-to-two driver/truck (or slightly larger) enterprises, as they rely mostly on themselves to plan and execute their daily objectives, the larger transport companies are set up to allow each professional to focus on their task at hand. Would introducing self-driving capable trucks eliminate the need for transport managers, for example?
Before the question could be put forward in a roundtable discussion, another topic has to be discussed – safety.
Safety of self-driving vehicles
While the hypothetical scenario of a driver, either professional or someone on their commute, having the ability to make business decisions and reply to emails, or to simply relax while a vehicle takes them to their destination is seemingly the one that’s very close to being fulfilled, it has to be met with some skepticism.
Researches at the Massachusetts Institute of Technology (MIT) looked at how attentive drivers were following the activation of Tesla’s Autopilot system, which is classified as an SAE Level 2 feature. While Mercedes-Benz’s Future Truck would be an SAE Level 4-equipped vehicle, the basic functionality of the computer taking over the active driving is the same. It has to be pointed out though, that per the SAE classification, Level 4 is named as High Automation, and the chances of a human driver having to take the wheel to avoid a potential accident are very low under that level of automation.
Nevertheless, the roads will continue to be filled with many ever-changing factors, from weather to independently thinking drivers, and as SAE defines it, Level 4 is only able to provide fully automated driving features “under limited conditions and will not operate unless all required conditions are met.”
MIT’s scientists concluded that “changes in glance duration and pattern suggest lower visual attention to the forward road when AP was engaged compared to after the disengagement to manual driving,” as most of the looks that were not directed at the road “were presumably non-driving related, as they were directed downwards and to the center stack region.” Conclusions should be drawn cautiously, as “off-road does not, however, automatically imply distraction or inattention because driver behavior cannot be assessed in isolation of the driving situation,” noted the group of researchers.
Safety and attentiveness are several times more important when driving a truck, as the vehicle, simply put, is much larger. Thus, it has to be put into question could truck drivers, before vehicles on the road are fully automated, truly do not pay attention to the road and be busy with other tasks, perhaps even sleeping while an SAE Level 4 vehicle does the driving for them? After all, at which point in time does the driver become the redundancy for autonomous vehicles to prevent a crash?
AI-based tools in other areas
The cabin is not the only location where AI technology could help road freight transport companies make strides in being more efficient and cost-effective, as moving the cargo is just one part of the whole road transportation process. The sales, planning, and executing deliveries are crucial to even begin moving any kind of cargo.
Introducing Artificial Intelligence into the daily operations of a road transport company is no easy feat, however. One of the key factors is ensuring that the data that feeds the software solution is up to par and presents a realistic picture, as a missing link could result in a chain that is unable to handle the whole load, leading to cracks and inefficiencies across multiple processes.
“The basis of the integration of AI-based technologies into our process has been our work behind the scenes, as together with our partners, we firstly looked at our back garden, before we decorated the front of our house, so to speak,” commented Pavel Kveten, the Chief Operating Officer of Girteka Logistics European Business Unit. “While installing new processes and systems can look straightforward from the outside, working with my colleagues made me understand how difficult and valuable that work is,” he added.
The logistics business, including road transport, has many variables present daily. Such factors, ranging from harsh winter weather, unexpected traffic along the highway, to a driver’s available working hours can shift the way that a customer’s delivery is handled.
Currently, according to Kveten, there are more than 200 possible factors that could influence a single truck’s daily plan that Girteka Logistics’ transport managers have to monitor continuously.
“Having an AI Operator, for example, helps us tremendously to be more efficient all across the board, as we can fine-tune every journey and every delivery, including the most optimal route to take, the optimal location, and time to take a rest, etc. From the driver’s perspective, doing everything in our Transics transport management system (TMS) eases their workload, as they do not have to input addresses across several navigation providers, and they do not have to look for parking locations manually,” continued Kveten.
Further inclusion of Artificial Intelligence
The AI operator is not the only AI-based tool used by the road transport provider. The AI Planner is another software solution that has been implemented into the company’s daily operations, as the planning application helps to oversee that trucks are connected with loads efficiently, timely, and most importantly, adhering to the customer’s requirements. If a customer wishes a two-driver truck for his delivery, the AI Planner can adhere to that requirement – only if that requirement is visible by the system itself, affirming how important the correct data is to such software.
“While we have begun to utilize AI-based software for our operations relatively recently, we saw an immediate positive impact to our efficiency. After all, we have to manage several thousand trucks daily, with a single decision potentially affecting our whole fleet of vehicles, thus, we have to be very meticulous in how we approach planning and operating our 9,000 trucks,” stated Kveten.
The future is bright for AI-based tools, noted the executive. “It will help organizations utilize their resources much better, allowing supply chains to be more resilient, efficient, and even more sustainable, as fewer resources will be wasted – such as by reducing the number of empty kilometers,” he concluded.
Perhaps the usage of Artificial Intelligence is still in its early infancy in logistics, as firms are looking into what kind of solutions would suit their needs the best, with self-driving trucks still seeming like a very distant future. However, with manufacturers already deploying various safety systems, such as lane keep assist or automatic emergency braking, the industry is seeing the first baby steps towards an automated future on the trucks themselves, while businesses are slowly deploying advanced software solutions in their processes. Predicting a future where not only business processes will be assisted by the help of software but vehicles carrying customers‘ cargo will be driven by an onboard computer with little-to-no human intervention is not a question of if but rather of when are the technology, the industry, and regulations ready to fully accept AI-based frameworks into their daily operations.