Predictive Analytics

Transforming business operations using a data-driven approach to create solutions

The client is a mining company based in Canada that wants us to reduce the cost of transportation delay. Now this delay may come from basically two reasons: –

Automotive reasons, like the failure of the parts of the trucks that directly leads to the delay of the product.
Humane reasons, like the driver being careless and not driving properly or going down the wrong path.
To counter these reasons, we have tried to create a mobile app /web page using models and codes applying Machine Learning Practices.

Technical Video

Features of the App

Predictive Maintenance

Driver Assist

Route Optimization Feature

The Wireframe of the App

Mobile app wireframe

Web app Wireframe

Predictive Maintenance


This feature allows us to predict if any mechanical/electrical part is going to fail and notify the driver. Using the data coming from the sensors of trucks, we try to predict when a certain part of the vehicle is likely to fail using a regression model. This way we alert the user to take a look at that part of the vehicle.

Part Replacement

Look for vehicle maintenance services on route/nearby and contact them for part replacement.  The app sends an e-mail notification to the driver regarding the contact info of the nearest shop where the driver can get a replacement.

Remote Switch

This feature ensures that control automatically switches between primary and secondary components if one fails.

Driver Assist

Driver Distracted

We know that drivers should always focus on the roads, else the chances of accidents shoot up exponentially. So, we have tried to apply this using a CNN(Convolutional Neural Network) that takes real-time data from the camera about the driver and predicts if he’s focusing on the road or not. If he isn’t the app warns the driver to focus on the road.

Driver Drowsiness

Some sources state that  72,000 crashes, 44,000 injuries, and 800 deaths in 2013 were caused due to drivers sleeping while driving. Now, this is not only terrible for the company but there’s also a life at stake. We try to take real time data from the driver and predict if he’s sleepy. If he is, we alert him via the app.

SOS signal sender

The model would send an SOS signal to the nearest Hospital/Police station if the driver seems unconscious, he/she's bleeding profusely or there is a serious equipment failure that might have been caused by an accident.

Route Optimization

Traffic Route Optimization

The best way to have the delay reduced excluding the automotive and humane failures is to have the driver drive the trucks through the best possible road to reach the destination fast.

Severe Event Forecast

Check for status and alert drivers about oncoming storms or unplanned gatherings. Look for events on Twitter using Tweepy and alert if they count for keywords such as 'hurricane', 'storm' or 'protest' are more than usual.

Rerouting in case of driver drowsiness

A drowsy driver is a danger, so we thought he should recommend some places where he might want to rest for a bit. Though, he would be late, but it’s better than losing a life.

Contact us for a quote, help, or to join the team.



(732) 347-6245

About Us

iSmile Technologies is a global technology services company.
(732) 347-6245


+1 (732) 347-6245
241 Jonathan Way
Bolingbrook, IL 60490


2-3-285, Secunderabad Hyderabad 500003


3190 Stocksbridge Ave
Oakville, ON L6M 0A7