Data disruption in logistics

(Photo: Shutterstock)

(Photo: Shutterstock)

One of the biggest trends in the field of freight logistics is the digitization of the trucking industry, which has a market of over $700 billion every year. Traditionally, truckers write out their logs on paper, carriers audit those logs, which can be a logistical nightmare. Tracking drivers through phones can also be a problem with cellular connectivity issues.

Over 70% of all freight tonnage in the U.S. moves on trucks. Silicon Valley has realized the potential of the freight logistics business and is investing heavily in start-ups which work on automating the process. Keep Truckin, a trucking automation start-up from San Francisco, is helping truck drivers by giving them the tools to keep a digital log of the driving hours. Keep Truckin sets up its device on the truck and drivers can keep an account through a smartphone app and also let the trucking companies know the real-time location of the truck.

Another technology that is seeping in from the digital taxi hailing system is the idea of digital freight matching. Uber introduced its Uber Freight in early 2017, trying to disrupt the age-old system of endless direct negotiations. Trucking companies and independent operators can browse through the freight available for loading in their vicinity, the distance it needs to be hauled, and ask shippers for upfront payment - making the process instantaneous and highly efficient.

Making the payment upfront solves a bigger issue for the truck drivers. Before the advent of digital freight hauling systems, demanding payment from clients was an arduous task, with it taking nearly a month or more. Uber helps carriers get their money within a week and also pays extra if there’s an extended waiting time.

Apart from easing freight and truck connectivity, logistics management companies understand the value of big data derived from the trucking industry. Starting with suggesting the best possible route for freight hauling, they also use big data to understand fuel consumption and ways to improve efficiency.

Measuring various parameters over the journey of a truck could help optimize resources and reduce waste. Trucks, in general, consume a significant chunk of the total gas consumption in the U.S., with over 38 billion gallons of diesel consumed last year. Reducing that usage can reap enormous benefits. Truck platooning is one way to make that happen. Platoons link two or more trucks in a convoy by using digital connectivity and helping them to maintain a specific distance for reducing drag and optimizing fuel consumption.

Peloton is an automated vehicle technology company which specifically works on truck platooning and thereby tackles the issues of fuel use and crashes on motorways. Its collision mitigation systems keep trucks safe, and radar sensors detect 360-degree vehicle activity and can apply brakes when needed. In the case of a potential platooning ahead, trucks are alerted and once in the zone, can be connected to vehicle-to-vehicle (V2V) communications.

Artificial intelligence models have transformed big data collected into tangible insights for trucking companies. UPS uses its data to understand fuel logistics. UPS delivery vans don’t necessarily take the shortest transit paths between destinations, and they also don’t turn left unless necessary because UPS routing software includes a policy that makes sure their drivers don’t turn into oncoming traffic. Though this might sound odd, it is a decision based on mathematical models and statistics.

Vans typically take only around 10% of left turns on their journey. And even if the van might be going in the opposite direction of the destination by not making a left turn, it still ends up conserving fuel and reducing the possibility of accidents and delays. This particular technique has saved UPS over 10 million gallons of fuel and helped it deliver 350,000 more packages per year.

Big data is also used to predict the delivery times more accurately, by taking the historical data inputs of environmental models. For example, during harsh winters, the time required for hauling freight over long distances increases significantly, and this could be predicted through data analytics. This improves the time duration estimation, and thus the delivery logistics become more manageable and controllable. Sensors fitted on the truck can also detect signs of wear and potential breakdowns and alert truck drivers about when to take the trucks to the shop. Providing leeway for such maintenance also improves the accuracy of delivery time promises.