The trucking industry has endured unprecedented upheaval over the last few years, largely driven by the confluence of demand volatility, macro-economic events, large-scale disruptions, call for sustainable modes of transportation, and rising customer expectations. Juxtaposed with the rise of e-commerce and increasing demands on final mile fulfillment, providers are under immense pressure to adapt their operations and thrive in the new normal. Despite the relentless deluge of challenges, the trucking industry remains an essential part of the global economy, and its continued evolution will play a key role in shaping the future of global trade.
In recent years, technology advancements such as automation, connectivity, and data analytics have provided opportunities for the industry to make genuine breakthroughs. However, the sector being a laggard in digital fitness is jeopardizing the adoption. Notably, new technology-driven entrants are innovating to fill the gap and carve out their niche in the value chain at a breakneck speed. One area at the forefront of this innovation is artificial intelligence (AI). AI has the potential to revolutionize all functions of supply chain management by enabling organizations to optimize processes, reduce costs, and increase efficiency.
This article is part one of a two-part series on how AI can address and alleviate major challenges in the trucking industry. In this part, three challenges are detailed with the AI-driven solutions highlighted.
Driver hiring and retention
The driver shortage has been a chronic challenge for the freight industry since the great recession of 2008. The American Trucking Associations (ATA) projects the shortage to skyrocket from 60K in 2021 to 160K in 2030. With the influx of new drivers entering the profession not compensating for retirements and voluntary career changes, retention and maximizing the productivity of existing drivers has assumed paramount importance.
Route planning and driver scheduling using AI can improve utilization and free up drivers to focus on the driving aspect of their job. Hesitant early adoption has given way to the avid embrace of advanced driver assistance systems (ADAS) to detect and prevent accidents, monitor driver performance, and alert issues like fatigue to improve the safety aspect of driving. Analytics on factors causing driver resignations can improve employee experience reducing turnover in the long run. In an industry that not long ago relied on paper-based processes and newspaper advertising for recruitment marketing, drivers are now interacting with NLP chatbots and simulation-driven tools to assist everywhere from recruitment to onboarding to routine driver support. “AI is appearing in nearly every area of the driver’s world and with technology has become a key value proposition for our drivers,” says Mike Hayden, SVP of Transportation Services at NFI Industries.
Demand and supply matching
In the last five years, trucking capacity in the U.S. increased at a rate of 1% while tonnage has increased at a rate of 1.2%, exhibiting significant fluctuation in magnitude and geographical distribution. To position capacity where demand is, carriers and owner-operators leveraged load boards when they could and deadheaded otherwise. In spite of their proliferation in the spot market, fragmentation, limited visibility, and manual processes limit load boards’ appeal and effectiveness. Interestingly, it is widely estimated that one-third of all trucking miles are run empty.
Digital freight matching is an AI-enabled capability that streamlines the shipment journey through end-to-end digitization. The load-capacity matching process is automated based on shipment requirements (O-D pair, truck type, insurance, service) by machine learning, tendering process automated by communicating technologies (EDI, API), and real-time visibility during execution facilitated by tracking. AI can also predict the future demand facilitating capacity positing to match the demand and design continuous moves based on available loads. The improved matching of available capacity to demand maximizes asset utilization, minimizes empty miles, and improves capacity repositioning.
Dynamic ETA
Estimation of a truck’s expected time of arrival (ETA) is a complex endeavor influenced by loading delays, traffic, weather, road closures, breakdowns, and human error causing a cascade of ad-hoc actions and suboptimal performance. The accuracy of the prediction is crucial in many steps of the downstream processes, such as managing appointments for the yard, unloading at the dock, allocations, fulfillment of the inventory, labor assignment in the warehouse, and execution in a busy production unit. While delays are inevitable, improving the accuracy of ETAs is critical to maintaining efficient flow of product.
While systems are still evolving towards real-time updates, artificial intelligence (AI) is helping to ease this transformation by dynamically calculating ETAs, using a scientific approach to produce real-time predictions through complex algorithms. AI uses various live data inputs including, but not limited to, historical pattern tracing, GPS information, onboard computer (OBC) data, driver data, EDI updates, equipment performance, and maintenance information. Since the data is so specific, AI models result in more accurate and more relevant predictions, adjusting for different operational locations based on inventory patterns, drivers’ behavior, truck timing, and geo-location data, etc. As the season or pattern changes, AI will adjust the model and provide realistic predictions and calculations that keep improving by continuously learning from the stream of real-time data.
“These dynamic, AI-based ETA calculations improve efficiency in receiving, accuracy in outbound fulfillment wave-runs, speed of store delivery, and e-commerce sales and customer experience,” says Mansoor Khan, founder and CEO of CSCS LLC (Cloud Supply Chain Solutions).
The future is bright with the impending breakthrough and commercialization of exciting advancements in autonomous trucks, internet of things (IoT), drone-based fulfillment, and blockchain.
About the author
Ashok Viswanathan is the director of digital supply chain and analytics at Best Buy and an adjunct professor at the Rutgers University Business School.