Vivid images in the media illustrate how climate change and severe weather patterns can wreak havoc on logistics. But this scene is about to change, with innovative artificial intelligence (AI) solutions for transporting goods more efficiently. Real-time fleet management insights, smart truck routing, optimized logistics networks and well-controlled hours will be shaping a new landscape for road warriors.
Without the benefit of big data solutions, product delivery is far less efficient than it could be. This can lead to delays and dangers for truck drivers—and commuters who share the roads. Each interruption can send transportation costs higher, create supply chain shortages and eventually take a bite out of consumer wallets. And idling trucks just add to greenhouse gas emissions.
Going full speed ahead to investigate possibilities for harnessing the power of AI for safer and more efficient trucking are seasoned AI scientists at the National Research Council of Canada (NRC) and the University of Calgary (U of C).
AI collaborations to the rescue
According to Program Leader Margaret McKay, the NRC's AI for Logistics Supercluster Support program "encourages the development of AI-driven solutions suited to regional conditions and realities across Canada." She adds that the most effective approach to devising successful systems that truly "understand" local conditions is to strike R&D collaborations that engage local stakeholders.
To craft such solutions for their Canadian partners, the NRC and U of C are working with the City of Calgary and 2 of the country's largest transportation companies, Canada Cartage and Bison Transport. AI tools developed during the 3‑year, $1.5‑million studies will help these collaborators optimize route planning and delivery schedules, improve supply-chain decision-making and enhance workforce planning. In the end, they hope to save time, money and fuel while increasing road safety and acquiring tools that can be adapted to new—even unforeseen—challenges. And the solutions will help smooth the way for Calgary commuters who share the roads in blizzards, rainstorms and hail.
With support from these partners, the NRC and U of C teams have accessed and analyzed vast amounts of information to identify specific logistics roadblocks and opportunities for using AI to solve them. The City of Calgary supplied the NRC and U of C with operational data that includes traffic speed and flow, while Canada Cartage and Bison Transport provided warehousing and delivery data.
Dr. Yunli Wang, the NRC's lead project investigator and AI specialist, points out that her team's expertise in state-of-the-art AI methods, such as graph neural networks and reinforcement learning, are key. "The disparate data needs to be integrated, so we use machine learning to improve forecasting," she says. "This means linking Calgary's historical traffic counts and weather data to study the impact of extreme weather on transportation systems, and developing a tool to recommend optimal routes for transporting goods around the city."
The U of C's lead investigator and professor of Geomatics Engineering, Dr. Xin Wang, adds that she and her team of postdoctoral fellows and PhD students contribute knowledge in geospatial data mining and facility location optimization. "We looked at more than 22 million GPS records to identify traffic patterns and driver behaviour," she says. "This yielded information about how trucks are moving around the city at different times, which routes they take and so on."
Madhuri Seera, Acting Manager of Transportation Strategy, City of Calgary, explains that the tools developed by the NRC and U of C "will help us meet the goals of our Goods Movement Strategy—a road map to achieving a highly efficient network in the city as growth in the supply chain and logistics industry explodes." Truck movement on highways around Calgary has increased by 55% over the past 15 years. Logistics funds about 134,000 jobs in Calgary, contributing some $14.5 billion to the economy.
The NRC and U of C are also using machine learning to help Canada Cartage and Bison Transport more efficiently schedule trucks and drivers. AI tools can supply information to drivers to help them serve their clients rapidly, efficiently and at a lower financial and environmental cost.
When the rubber hits the road
The network-based nature of logistics makes the transportation industry a perfect candidate for adopting and using AI—and the market is ready: it is projected to grow from $1.2 billion in 2017 to $10.3 billion by 2030.
As we look to the future of transportation, we can see big changes ahead, including AI-enabled driver safety tools. Moreover, rising fuel costs and the race toward carbon neutrality are calling for smarter routing, fewer empty-truck miles and reduced time on the roads.
Building Canadian AI talent is integral to this future. "Students and post-doctoral fellows who have participated in the program are in high demand," says McKay. "We provide training in real-world problems along with opportunities to work with industry partners. This helps to attract and retain experts in Canada."
With these game-changing projects in western Canada, the NRC and its collaborators are enabling the transportation industry to run in a more seamless, interconnected way. And that's a good thing for all Canadians.