Making Semitrucks Safer and More Reliable with MicroAI AtomML™
21531
post-template-default,single,single-post,postid-21531,single-format-standard,bridge-core-1.0.4,translatepress-en_US,bridge,ajax_fade,page_not_loaded,,qode_grid_1300,qode-content-sidebar-responsive,qode-theme-ver-18.0.6,qode-theme-onetech,disabled_footer_top,wpb-js-composer js-comp-ver-5.7,vc_responsive
 

Semitruck Management with MicroAI AtomML™

28 Jan 21
Semitruck Management

Semitruck Management with MicroAI AtomML™

The Industry Need

The semitruck industry is one of the most heavily monitored industries in the United States. Every year, new technology is developed to help trucking operators comply with regulatory requirements. One area that could still be improved upon is maintenance.

Semitrucks are already filled with sensors that measure various states and performance parameters. This makes them ideal candidates for MicroAI AtomML™ and predictive maintenance. By feeding in time series data from existing sensors, as well as potential additional sensors, the combination of MicroAI™ Atom and predictive maintenance will allow drivers and fleet managers to detect—in real time– the early signs of impending maintenance issues.

 

The Solution

The first step to utilizing MicroAI AtomML™ and predictive maintenance is to select the appropriate sensors to provide the required data. Typical sensors for this use case would include (but not be limited to) the following:

  • Temperature sensors: Monitoring engine component temperatures using only raw temperature data would not be sufficient. Feature engineering would be required to enable the comparison of component temperatures to ambient temperatures.
  • Weight sensors: These sensors can be deployed inside the cargo trailer. In addition to using raw weight data, the developer should consider comparing various weight measurements to create deeper insight into how those weights are distributed. This will enable the AI engine to detect significant imbalances in cargo weight that could put the trailer under uneven levels of stress.

 

semitruck-microai-atom

Depending on the specific use case, a much more extensive list of sensors could be utilized to provide additional data.

The next step will be to create alerts that will be triggered in the event that a sensor is exhibiting abnormal behavior. Not only will MicroAI™ Atom be able to detect thresholds in the data that have been crossed, it will also be able to detect unexpected changes in the data that is being produced. For example, if temperatures are changing too quickly, or if the relationship between ambient and internal temperatures are different.

Automatic corrective behavior would be difficult to implement in a semitruck environment; however, alerts can be raised to the driver and fleet manager to have different components of the truck inspected. These detections and frequency of abnormal behavior would then be fed into the predictive maintenance program which would in turn produce a health score for the truck as well as an estimate for when maintenance should occur.

 

The Impact

Since most semitrucks already have an abundance of sensors and MCU’s on board, MicroAI™ Atom can be quickly onboarded within the environment. Once onboarded, MicroAI™ Atom can begin the process of embedding and training AI and machine learning algorithms onto those MCU’s. The combination of MicroAI’s MicroAI AtomML™ and predictive maintenance will allow semitrucks to increase efficiency by being able to more accurately predict when their trucks will require maintenance. The result will be semitrucks that are safer, more reliable, and more profitable.