Sensor Data and IoT: Sensors, Trains and Analytics

You have the data:  Why not use it?

That’s the question that my company, One Point Solutions, asked in our booth at the recent Big M Manufacturing Convergence trade show in Detroit.  We used this toy train armed with a 10-reading TI sensor to simulate an assembly line with end-to-end monitoring, reporting, predicting and acting upon sensor data.  While this was a simple demonstration it showed the power of what can be done with data you already have — or data you can start collecting using very inexpensive sensors.  Booth pic-choo choo-thumbnail

This was a great show that allowed us to meet many exciting and diverse companies that included 1st-tier auto suppliers, their suppliers, niche manufacturers, and consultants to all of the above.   It was great to hear how companies are using — and not using — sensors now.

 

The things I heard the most were:

  • “We collect sensor data but we don’t really do anything with it.”
  • “We know about sensors and are interested in finding out how they can help us.”
  • “We’re too busy with our day-to-day process but that sounds interesting.”

Well, it’s not as hard as it sounds and we can use a rapid prototype to provide a quick Proof of Concept for your company.  Here’s how we did exactly that at the Big M (also see our whiteboard):  wb-EVERYTHING-thumbnail

1) Purchased $30 sensor from TI, which provides readings such as temperature, humidity, gyro, accelerometer, and tilt (left, right, up, down), and attached it to the top of the train, then set up the track to simulate each reading (see labels on above picture).

2) Connected the sensor to an iPad and to the computer monitor in the background using its built in software and IBM Bluemix.  As the train moved down the “assembly line”, real-time data was displayed on the iPad and monitor (see pictures below). Booth pic-screen2-thumbnail3) Like in a real-world scenario, we set thresholds that might indicate a problem on the assembly line (in this case, we used tilt left and up).    When the thresholds were exceeded, a tweet was sent (lower left-hand corner of monitor) and text messages were sent, alerting shop-floor personnel of a potential issue.

4) The data was collected and sent to the cloud and a local database, where it can be analyzed and acted on using products such as IBM SPSS and IBM Predictive Maintenance and Quality.

PMQ-desktop-John

That was it!  Simple but effective.  Try extrapolating that scenario to your company’s train track.

Since I founded One Point Solutions in 1997, we have been all about the data. If you have it, and can tell us about it, we can provide solutions for you to store, analyze, and act upon it.

Find how to to try this now with your data >>>

– Ron Flannery,
President, One Point Solutions
248-887-8470 x 40
PR@One-Point.com

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s