The use of wireless technology for asset tracking has taken off over the last decade. Consumer retail was the first industry to adopt this technology for location services by tagging product pallets with radio frequency ID (RFID) tags. In 2005, retailers such as Walmart made it mandatory for all of their suppliers to use such tags to allow inventory tracking. This was followed by major automakers such as BMW and Toyota and shipping companies such as NYK Logistics adopting RFID asset management. The use of such tags for tracking human activity has also made some inroads – in patient care, in agriculture and in personnel tracking in hazardous environments. However there are no applications in manufacturing, particularly targeted at SMMs which are involved in contract, custom manufacturing operations that allow accurate tracking of employee time for various production related activities. Current RFID applications for labor tracking end at basic location services, i.e. pinpointing worker locations to enhance safety, for example. But there is potential to use this technology beyond simple location prediction, for example for accurately tracking labor costs.
Currently, location systems install RFID Readers at strategic locations. When people wearing the corresponding tags pass within the range of the scanners, the system is able to identify the location based on the tag ID and scanner placement.
Tracking or monitoring labor hours requires installing these scanners or sensors at various points on the shop floor, for example. The sensor can identify the tag and report that identification to a remote database. There are three main challenges to effectively using this technology for automated labor cost tracking. Unlike physical assets, people on a shop floor are not immobile all the time. This means that they may tend to wander into areas which are not completely covered by the sensors.
The first challenge involves correct placement of the scanners so that physical obstruction, interference with other RF emitting machines and detection range do not affect the signals.
Secondly, if a workstation sensor is close to an entry or exit point, we may detect several false positives based on employee movement. Algorithms need to be developed to manage the tracking by accounting for such possibilities. The second challenge involves analyzing this data so that effort is correctly attributed to the employees and work tasks.
The next challenge is handling the generated data. For continuous activity tracking, the sensors must validate proximity of the tags quite frequently – at least several times a minute. All of this requires resources to install, maintain and operate the hardware and accompanying database software. A typical RFID sensor unit alone can cost about $4000 and as the number of work zones increases, the cost and set up complexity can quickly add up. Sensor vendors supply an accompanying data warehouse to store the generated data which can also add server and related IT costs. Depending upon the number of scanners and employees, this data can quickly add up to unwieldy levels for most SMM IT capacities. This complexity and resource requirement has currently limited the large scale adoption of RFID for labor cost tracking.
The biggest promise that future holds comes down to pricing and adaptability. The three challenges listed above are fundamental issues which have to be dealt with. However IoT based technology holds the promise of dramatically reducing the price of the hardware and making the communication between the sensors and source of signals far more flexible. The cost of the signal producing hardware will drop down to a few dollars and the receivers of these signals would be the most basic smart phones or tablets whose prices will also drop to the low hundreds of dollars. But the underlying data and analytics issues will still have to be dealt with.
Image courtesy: beekn.net
Bala Deshpande, Founder, SimaFore
Dr. Deshpande’s has 19 years of experience in using analytical techniques. His first exposure to predictive models and analytics was in the field of biomechanics – in identifying correlations and building multiple regression models to predict muscle forces based on electrical activity in muscles. He began his career as an engineering consultant at EASi Engineering, following which he spent several years analyzing data from automobile crash tests and helping to build safer cars at Ford Motor Company. He has been instrumental in promoting information theory based analytical techniques for a range of applications from performance measurement in organizations to predicting patient stability in ICUs. He holds a PhD in Bioengineering from Carnegie Mellon and an MBA from Ross School of Business (Michigan).