Accurately projecting and communicating shipment details in freight and logistics is paramount for successful, safe delivery across the vast supply chain.
Timely and precise information regarding shipments’ whereabouts, expected arrival times, and potential delays is crucial to ensure efficient operations, optimize resource allocation, and meet customer expectations. Late or delayed shipments can have severe downstream impacts, disrupting manufacturing processes and causing significant financial losses.
Yet despite the fact the industry has long been inherently replete with data, the logistics landscape remains stuck mainly in outdated communication practices and manual processes — from a technology adoption standpoint. The reluctance to evolve in logistics has hampered efficiency, increased costs, and leaves room for errors and delays. There is a pressing need to embrace technology and automation to overcome these limitations and drive progress in the industry.
Dr. Yossi Sheffi, an expert in global supply chains, emphasizes the importance of total transformation of the industry to leverage AI and its potential to analyze vast amounts of data, stating that “even with current available AI tools, it’s hard to see past that third degree of relationships between supply chains.”
While AI can provide powerful insights, its effectiveness relies on the quality and availability of data. Only with the right tools and processes in place can AI help effectively predict outcomes and enable proactive decision-making.
Daniel Faggella, CEO and Head of Research at Emerj Technology Research, recently sat down with Convoy CTO Dorothy Li on the ‘AI in Business’ podcast to explore the possibilities of leveraging the massive amount of data stored in freight and logistics. They discuss how AI can accurately predict and plan for freight shipments, bringing greater efficiency and transparency to the industry.
From their conversation, this article derives the following listed key insights applicable to enterprises optimizing any value chain with AI and machine learning:
- Proactively detecting driver delays: By incorporating real-time Google Maps traffic data in training machine learning models, businesses can accurately predict the probability of delays and take corrective action.
- Democratizing data to increase logistics transparency: Even smaller players across the supply chain can access the same level of information traditionally reserved for prominent industry leaders.
Listen to the full episode below:
Guest: Dorothy Li, Chief Technology Officer for Convoy
Expertise: logistics, supply chain, data analytics, machine learning, artificial intelligence, digital transformation
Brief Recognition: Before serving as CTO for Convoy, Dorothy spent more than two decades at AWS, where she helped launch some of its most important innovations, from Kindle to Amazon Prime.
Proactively Detecting and Mitigating Shipment Delays
Traditionally, logistics companies have based their freight shipment predictions on historical data and human judgment. Unfortunately, these estimates have often fallen short due to the complexity and variability involved in freight shipment. Weather conditions, traffic patterns, and a multitude of human factors can all dramatically influence a shipment’s timeline.
Especially in the world of long-haul trucking, the age-old practice of manually coordinating details for every shipment is laborious, time-consuming, and vulnerable to human error. Unavoidably human-centric, these practices often involve repeated back-and-forth communication through phone calls or emails, even to confirm the schedule or status of a delivery.
Perhaps most problematically, the variety of manual methods that have traditionally been employed lack scalability. As logistics operations continue to grow, the ability of logistics companies to manage the growing array of shipments relying on methods that long preceded the advent of digital technology has become an increasingly daunting, if not impossible, task.
Fortunately, the advent of artificial intelligence and machine learning has opened up the possibility for an entirely new era of reliability and speed.
Successfully entering this new era not only requires technologically-adept leadership to realize its potential, it also involves a comprehensive overhaul of processes, people, and paradigms within the organization, right on down to the individual trucker: “It may or may not be surprising to the audience that there’s very little visibility in supply chain logistics today,” Li tells Emerj.
“Most things are still very manual, there’s no GPS tracking, and the truck, the driver, most of them – unless they’re actually using Convoy – they don’t use a mobile app, which means that there’s no tracking,” she continues.
At Convoy, Li explains that today, every driver is given a mobile phone with enabled GPS capabilities and the Convoy app included. She tells Emerj the decision was initially controversial, given the importance of truck drivers keeping their eyes on the road. Still, these days, more than 95% of Convoy employees are fully compliant.
“We’re now able to track almost 1,000 data points on every shipment, as well as up-to-the-minute ETA for the truck driver, so we know exactly where that shipment is,” Li says.
The data that is shared isn’t limited to one-way communication, however:
“The driver can also give feedback to the shipper to the facility, so sometimes if the facility is really congested, or if they have an issue in the facility, they can let us know, so that also helps improve the shipper’s operation and shippers tell us that they really appreciate that part,” Li adds.
Democratizing Data to Increase Transparency in Logistics
As the logistics industry continues to evolve, another critical area that will see significant progress over the next several years is the improvement of data infrastructure and the digital transformation of smaller companies. Because of this, transparency and visibility across the entirety of the supply chain will increase to benefit large and small companies.
In the traditional world of freight and logistics, even in the wake of widespread digitization across most other sectors, only large brokers had the access and technology to allow for the level of transparency needed for the industry’s more extensive digital transformation.
However, advancements in technology, including IoT, increased mobile phone usage in the trucking industry, and the availability of real-time data streams have democratized data access. Even smaller players across the supply chain can now access the same level of information that was once reserved for industry leaders.
Li describes that at Convoy, because every driver is provided with a mobile phone equipped with GPS as well as the Convoy app, it is possible to track shipments in real time, enabling better visibility throughout the supply chain.
This democratization of data in the freight and logistics industry is ushering in a caliber of transparency that is leveling the playing field for all stakeholders.
Shippers gain visibility into their shipments’ status and precise location, enabling better resource planning and allocation. Independent drivers and smaller trucking companies can leverage valuable insights to optimize routes and unearth new opportunities. Naturally, customers benefit from real-time updates on their deliveries, enhancing their overall experience and level of satisfaction.
Moreover, this democratization of data brings about a more collaborative and efficient ecosystem for the freight and logistics industries. As Li explained, the shared data allows drivers to provide feedback to shippers about facilities, which makes for improved operations and streamlined processes. Issues such as congestion or facility-related challenges can be promptly addressed or, even better, circumvented altogether.
As such, the entire logistics industry stands to gain from increased transparency, efficiency, and collaboration as smaller trucking companies leverage technology and access valuable insights.
Furthermore, integrating AI in freight and logistics opens up avenues for new business models and partnerships. As companies gain access to real-time data and predictive insights, they can explore innovative ways to collaborate and optimize the supply chain. For example, AI-powered platforms can facilitate dynamic matching of available capacity with shipments, enabling more efficient utilization of resources and reducing waste.
This shift towards data-driven collaboration and optimization not only benefits individual businesses but also has the potential to transform the industry as a whole: Businesses can build trust and loyalty among their customers, leading to improved satisfaction and long-term partnerships. Furthermore, AI can assist in predictive maintenance, allowing logistics companies to identify potential equipment failures or maintenance needs proactively. By addressing these issues before they cause disruptions, businesses can minimize downtime and ensure the smooth flow of operations.
In addition to optimizing day-to-day operations, AI has the potential to revolutionize risk management in the freight and logistics industry. With access to vast amounts of historical data and sophisticated algorithms, AI systems can analyze patterns and identify potential risks or bottlenecks in the supply chain. This proactive approach to risk management enables businesses to mitigate potential disruptions and develop contingency plans in advance. By leveraging AI, companies can enhance their resilience, reduce the impact of unforeseen events, and maintain a competitive edge in an increasingly complex and volatile global market.
However, as Dr. Sheffi notes, even with the most advanced technology on the market today, without the proper quality and quantity of input data, such technology is of little use.
With regard to the technology available, Dr. Sheffi notes, “It’s not the AI – the API is there in terms of analyzing data,” he says.
“The main reason is getting the data, and getting the data is still not there; even with the internet of things and having sensors all over the place, there are some areas where you still can’t get the data.”
