Shoot for the moon: Pushing the boundaries of transport sustainability and performance
Just over 50 years ago, we landed on the moon. What is perhaps humankind’s greatest achievement was accomplished with more than a decade of research and development, warehouses full of calculations made with pen and paper, and less computing power than the average person has in their smartphone today. In the decades since, the rapid development of technology has resulted in an exponential improvement in capability, and as a result, the recent NASA-SpaceX mission was largely automated, a far cry from the comparably primitive space capsules of the late 1960s.
In contrast, the transport networks we rely on today to move goods and supplies on the road are nearly as analog as ever. Rudimentary software systems and pen-and-paper solutions remain the norm, coordinating freight that is carried out by unsustainable diesel trucks. In fact, according to 2016 data, road freight transport vehicles in Europe covered the distance to and from the moon over 176,000 times in a year. As a result, the road freight industry is responsible for 7% of all global emissions and expected to increase that figure by an additional 15% if nothing changes. Solutions to this problem exist and are being rapidly developed today — specifically electric and autonomous freight — but the question remains: if we’re shooting for the moon, why not make use of technology to get there as quickly and easily as possible?
Electrifying and automating road freight may not be as complicated as space travel, but there are copious unique variables to consider and plan for that will make the solution exponentially more effective. The best way forward for road freight to eliminate emissions and reach global targets while remaining cost-competitive with diesel freight is through the wide-scale implementation of autonomous electric transport (AET). That means implementing networking software for electric and autonomous transport that considers the unique characteristics of these new freight systems.
Accounting for countless electrification variables with machine learning
Operating electric vehicles (EVs) and ultimately AET requires a coordinated approach to vehicle planning, monitoring, and integration, with variables specific to a particular transport flow. Electric trucks have unique considerations that make them much different to operate than diesel vehicles, and as such warrant a different approach. Changing this requires thinking beyond the capabilities of current freight networks to what is possible through technology today.
We designed the Einride platform not only to link all vehicles operating in the fleet digitally, but to provide optimized, dynamic route planning based on shipping demands, loading dock availability, and other considerations unique to electrified transport. The intention is to carefully manage the transition from diesel to electric and autonomous vehicles on a massive scale, and as soon as possible. Through machine learning, the more data the platform receives, the more effective it becomes at identifying bottlenecks and coordinating individual transport networks for the unique scheduling, routing and infrastructure demands of electric freight.
In order to realize reductions in a freight network’s overall emissions and improve cost-effectiveness, the platform requires tracking and analyzing historical routes, driven distances, shipping volumes, and other relevant data. With this information, it can update scheduling, routing, and other protocols to account for relevant variables in a given network. Subsequently, it will highlight areas where electrification would have the most impact, and outline exactly how to make use of existing EV charging infrastructure as well as plan for new additions.
As an example of how this works in practice: one would expect that a 100km delivery between two points can be accomplished by an electric truck with 180km of estimated maximum range. However, that expectation fails to take into account the myriad of variables that can affect this range while still managing to maximize vehicle utilization and minimize operational costs. With a full load on board, topography variations, traffic congestion, weather conditions, and even whether the driver is using heat or air conditioning in the cabin, actual range can vary significantly. The platform takes all of these factors into account to provide not only detailed live network insights to both supervisors and drivers, but also optimizes charging, shift planning, and loading and unloading schedules for maximum efficiency and utilization.
In short, our solution transforms electric freight from a guesswork replacement for diesel trucks to an optimized network of emissions-free transport. But perhaps its most important function is to set the framework for a future of freight that is both autonomous and electrified. If widespread electrification is like getting to the moon, automation is like landing on Mars.
Setting the foundation for an autonomous electric transport network
Preparing for the future of road freight — one that is entirely electrified and autonomous with vehicles like the Pod — requires thinking beyond the limits of analog computational capability. By using the platform to digitize and process a large set of relevant data on transport electrification and using machine learning to improve its implementation, we can set the foundation for a smooth transition to autonomy and the additional unique infrastructure and considerations it will require.
The digital transport orders and facility update recommendations the platform generates serve as a basis upon which to automate loading, unloading, and charging activities, necessities for the implementation of autonomous freight as no driver will be on board to supervise. Without it, it would be difficult for shippers and carriers to coordinate autonomous electric trucks with the precision that the platform enables, even if they own and operate them already.
Beyond this, the platform is designed to scale functionality for the implementation of systems where one operator oversees and can control multiple vehicles at once. In the not-so-distant future, fleets of unmanned Pods will operate shipping routes autonomously, thereby increasing the freight capacity of an entire network while significantly reducing the operational cost and eliminating emissions. To reap the benefits of this system in the future, transport processes must be digitalized as early as possible, not just for autonomous systems but for electrification as well.
In transforming road freight to be not only sustainable and cost-effective but also scalable, we face a collective challenge that will require more than pen and paper to solve. Utilizing the technology we have available that is specifically suited to the task means implementing intelligent solutions that make the transition to autonomous electric transport significantly simpler and exponentially more effective. This way, we’re not just shooting for the moon, but pushing the boundaries of what is possible far beyond it.
Originally published at https://www.einride.tech on June 9, 2020.
