Posted By Jeff Moad, July 22, 2016 at 3:12 PM, in Category: Transformative Technologies
Manufacturers invest lots of money and intellectual capital defining and documenting the philosophies and standard practices under which they want their plants to operate. Operations bibles such as the legendary Toyota Production System not only lay out standard operating procedures aimed at driving up quality and productivity, they also reflect the cultural aspirations of the manufacturing enterprise.
But how can manufacturing leaders be sure plant operators aren’t deviating from or even ignoring their companies’ standard production system, possibly leading to quality problems or even costly recalls? To avoid this possibility, we see manufacturers emphasizing clear communication, training, employee engagement and, of course, supervisor oversight.
But is there a role for technology in making sure that operators are adhering to standard work and documented operating procedures? Technology conglomerate Hitachi Ltd. is attempting to make the case that there is. The company, in collaboration with one of its key manufacturing customers, recently announced plans to test a system that analyzes surveillance images of plant operators to allow for “rapid discovery of defects in machines and materials and deviations in worker activities.”
Hitachi said the system revolves around an image analysis system that is able to detect worker activities and movements that depart from established standard best practices. Image-based data compiled by the new system will be combined with manufacturing execution system information about machine and material performance to give manufacturers the ability to continuously monitor all aspects of production and to view results down to the individual product serial number, Hitachi said.
Hitachi predicted this approach will be effective “in improving product quality and in increasing productivity and the accuracy of traceability.”
The company said the system derives desired, standard worker behavior models based on “frontline interviews and observations.” The system then isolates and identifies operator actions that deviate from standard behavior models to understand the correlation to defects and to aid in correction of operator deviations.
Hitachi said the system can also “detect abnormalities in the case of welding defects by combining voltage and current data from existing facilities with light-emitting element color analysis using high speed cameras.”
Hitachi said it has been conducting joint verification of the system with its customer, Daicel Corporation, a Tokyo-based chemicals manufacturer. The two companies plan to begin operation of the system in Daicel’s Harima, Japan, plant this fiscal year. That will be followed by a rollout in six of Daicel’s overseas plants, the companies said.
Following testing, Hitachi plans to make the system available to the manufacturing industry as part of its Lumada IoT platform.
The idea of intelligent surveillance systems monitoring operator behaviors and movements on the plant floor raises a number of important questions. While employees would likely have little standing to pose privacy objections to use of such systems, it’s worth asking what impact plant floor surveillance would have on ongoing attempts to build employee engagement and process ownership.
Also, what sort of reaction would organized labor have to the use of such a system?
And how would such a system change the role of front-line supervisors on the plant floor?
If systems like Hitachi’s pan out and come to market, we may soon learn the answers to those questions.
Written by Jeff Moad
Jeff Moad is Research Director and Executive Editor with the Manufacturing Leadership Community. He also directs the Manufacturing Leadership Awards Program. Follow our LinkedIn Groups: Manufacturing Leadership Council and Manufacturing Leadership Summit
Unfortunately, too many plants don't run on the basis of clear, shared performance goals, as well as a shared stake in the improved financial performance. These plants, typically with low levels of engagement and trust, are likely to see the surveillance technology as just further evidence that leadership does not trust them. Their ability to foul up this system is better than most would believe. They can probably remember failed new technology implementations.
Creating a performance oriented plant with a focus on ongoing improvement was captured in this Forbes article, http://www.forbes.com/sites/fotschcase/2016/06/15/how-us-manufacturers-can-compete/ I suggest creating a performance focused environment with plant employees before implementing the next new technology.