machine learning

How can the Internet of Things apply to the Trucking industry - I’m glad you asked...!!!

A colleague just recently attended the North American Commercial Vehicle (NACV) fleet technology show, and sent me pictures of the 2020 model commercial trucks.

A colleague just recently attended the North American Commercial Vehicle (NACV) fleet technology show, and sent me pictures of the 2020 model commercial trucks. While it might surprise some folks given my role at Metal Toad, I'm a huge fan of heavy equipment and semi trucks. It probably stems from growing up working on a farm operating heavy machinery, and then being a Master Driver in the US Army. There is something intoxicating about being in control of that much power. 

As a result, I’m always thinking about ways we can help improve the transportation industry. The transportation industry is vital to economies, commerce, and honestly day-to-day living. In fact, roughly 71% of the freight moved in the U.S. is moved by trucks. With Thanksgiving rapidly approaching, stop and think for a moment “where did all of my groceries come from?” The quick answer: it was more than likely brought to your local store by a commercial truck. 

Reducing the costs of operations are the ‘moving goal posts’ for fleet managers and owner operators. At a glance the operating expenses (OpEx) break down like this: Fuel 39%, Salary 26%, Truck and Trailer 17%, Repairs & Maintenance 10%, Tires 3%. 

With some “plumbing” or integration, I see three areas that can leverage the monitoring and predictive analytics offered by partners like AWS: 

  1. Fuel
  2. Repairs and Maintenance
  3. Tires

Use Case Opportunity: Manufacturing systems that integrate with technology 
Imagine fusing telematics with terrain models to optimize fuel consumption. With the introduction of new transmissions (like the automated manual transmission from Paccar) smart shifting coupled with terrain models can now start to optimize engine performance based on grade, vehicle and load weight to improve fuel consumption during operation. In everyday speak, when using cruise control the truck knows when to speed up and slow down for climbing and descending mountainous terrain, and I know when to downshift and slow down for sharp curves and turns, etc.  There is hope — research in this area has been done at the Linköping University in Sweden using Volvo I-Shift and I-See intelligent cruise control, which has seen a reduction in fuel consumption of up to 5%. So it can be accomplished. 

OpEx summary for fuel

  • The largest single cost in trucking is fuel. 
  • The average price for diesel, as of Nov 18th, is $3.074/gal with the highest being $4.003/gal in California. 
  • Fuel tank capacity on trucks ranges from 125-130 gallons.
  • The average mpg for semi trucks is between 4-8 mpg, and can drop to 2.9 mpg going up hill!

Use Case Opportunity: Predicting optimal smart routes
Visualize taking historic drive times and traffic patterns and coupling that with current weather conditions, current traffic patterns and a driver’s behavioral operating patterns and apply Machine Learning (ML) and Artificial Intelligence (AI) to offer optimal start times and truck friendly routes to ensure a safe and on-time delivery. In fact, we could also layer in additional algorithms to start building an artificial neural network for Deep Learning that could include load boards that could evaluate the closest available loads to reduce wait times or deadhead miles (empty or no revenue) and the amount of driver hours remaining on their electronic logging device (ELD) to stay in compliance. Out-of-compliance fines from the Federal Motor Carrier Safety Administration range from $1-10k, with the average being just over $2800. 

Use Case Opportunity: Preventative maintenance through onboard sensors
Picture a scenario where telematics alerts and alarms from onboard sensors are being monitored, analyzed, and reported out in order to plan preventative maintenance. Let’s take tire inflation as an example, tire pressure monitoring systems should be a standard integration in navigation systems, similar to Peterbilt SmartNav or Aperia’s Halo Connect, but could also be expanded to include miles since last rotation, and remaining tread measurement for DOT inspections. A ‘state of the truck’ health dashboard would keep drivers and fleet managers better informed and could head-off issues before they become costly and time consuming maintenance expenses and delays. Detroit Connect works great and has the right idea, but is limited to detroit engines which are predominantly offered in Freightliners and Western Star trucks. It would be good to see this type of information integration reach more truck brands. 

OpEx summary for tires: 

  • The average semi-truck tire is $250, and you have 18 of them..!!!
  • Annual tire maintenance averages $4k, and that’s not replacing all the tires.
  • It’s been said that for every 10 percent of underinflation, you’ll see a 1 percent reduction in fuel economy. 
  • And, running a tire constantly underinflated by 20 percent will reduce tread life by 30 percent, and running at 40 percent underinflation will reduce tire life by 50 percent.

Admittedly, some of this should be normal operator preventive maintenance checks and services, but it’s not always done. As a former fleet manager, I would have loved to have a proactive dashboard to recognize potential issues and plan maintenance schedules as opposed to approving purchase orders for towing or roadside assistance! Being able to allocate budget dollars against predicted maintenance could also free up capital to invest in other profitable areas. 

Yes, IoT has a part to play in the transportation industry
The secret sauce to modernizing the trucking industry and helping improve efficiencies is: Integration or ‘plumbing’. Much of the data and technology capabilities are in currently in production, and once they’re integrated the actionable intelligence presented will allow the right decisions to be made quickly. 

In the works, and next on the blog agenda
Autonomous vehicles: Will California Assembly Bill 5 accelerate autonomous vehicles (level 4) in becoming potential point to point shuttle options? The work being done by industry partners like, TuSimple, Embark, NVIDIA and Paccar on neural networks and autonomous driving is very exciting.

Similar posts

Get notified on new marketing insights

Be the first to know about new B2B SaaS Marketing insights to build or refine your marketing function with the tools and knowledge of today’s industry.