Business drives Smart Manufacturing: Part 2
Transferring Human Knowledge to Machines
- By /
Ms. Sujata Tilak, MD - Ascent Intellimation Pvt. Ltd.
10 Jul 2017
In last article we saw various strategic and operational goals. Let us see ‘how’ these goals can be translated into action.
Reducing unplanned downtime is probably the most sought after goal by any shop floor manager. Unplanned downtime can occur due to many different reasons like machine breakdown, unavailability of raw material, unavailability of operator etc. Out of this, machine breakdown is most unpredictable, or is it really…...?
There are expert operators / supervisors / maintenance people in each manufacturing company who can predict impending problem much before it occurs. They have diverse ways of understanding machine health from sound, parameter variations, vibrations, current patterns…the list is endless. But these are experts we are talking about and every plant has only a few of this tribe!
Smart Manufacturing Systems assimilate this knowledge for Condition Based Monitoring (CBM). All the expert human knowledge about machine condition is configured into a rule engine or stream analytics module. The conditions are based on sensor data, time as well as occurrence frequencies. The rule engine uses sensor data to continuously evaluate conditions and raise alerts when any condition becomes true. Thus the expert knowledge is in action continuously for every machine. This is huge scaling of human knowledge into a Smart System.
CBM can be used for a variety of outcomes like
CBM can be fully operational on day one of installation and does not need any past data like Predictive Maintenance. Thus industries start getting returns from the system immediately.
From architecture standpoint, CBM can be distributed at the edge or centralized in the cloud or both depending on where and how the insights are turned into actions. It provides much needed flexibility and scalability to the system.
When we implement any PlantConnect solutions, CBM plays a major role in delivering tailor made system to each plant. For example, one of our PlantConnect RAMS customers has defined 31 conditions just to ensure best practices of machine usage are followed all the time. In an implementation of PlantConnect Insights in a dairy, 130 conditions are evaluated every second to execute entire business logic for reporting. In a PlantConnect SFactory installation, 17 conditions are evaluated every second to track availability and performance losses of a machine.
Find out more about our IIoT solutions on www.aiplindia.com