Computational creativity and the future of AI

Sensor sleeves could maximize workplace efficiency


January 5, 2012

Sleeves equipped with sensors have been developed to time the actions of factory workers, ...

Sleeves equipped with sensors have been developed to time the actions of factory workers, in order to increase efficiency

In factories where products are mass-produced, it's extremely important to know how long the human workers take to perform certain tasks. This not only allows the pace of the assembly line to be set, but it also allows factory owners to identify time-wasting problems such as superfluous movements, overly frequent tool changes, or impractically-located components. Typically, workers are periodically timed by a stopwatch-wielding supervisor, or using a timer that they start and stop themselves. A new wearable time-keeping system, however, promises more accurate readings.

According to Germany's Fraunhofer Institute for Factory Operation and Automation, there are at least two main drawbacks to existing time-keeping methods. For one thing, workers tend to be nervous when being monitored by someone else, and will work faster than normal, possibly also making more mistakes. Additionally, a supervisor timing their employees is being taken away from other duties, decreasing efficiency.

Fraunhofer's new system, designed under contract for engineering firm DR. GRUENDLER, incorporates two sleeves worn by the worker. Each sleeve contains three matchbox-sized sensors, located on the upper and lower arm, and the hand. These measure the acceleration and angular velocities of arms and hands in the X, Y and Z axes. After initially being "taught" by the user, the hardware can identify and isolate actions such as reaching, grasping, setting up, joining, checking or releasing.

The system doesn't require any extra infrastructure (unlike GPS), and allows multiple workers - wearing multiple sets of sleeves - to be timed simultaneously. Once the data has been gathered, a PC application reconstructs the motion sequences, breaking them down into precisely-timed individual actions.

Presently, the Fraunhofer system is applicable to assembly jobs at workplaces where employees are seated. Down the road, however, there are plans for it to be adapted for use with standing and moving employees, and for it to be able to detect their posture.

About the Author
Ben Coxworth An experienced freelance writer, videographer and television producer, Ben's interest in all forms of innovation is particularly fanatical when it comes to human-powered transportation, film-making gear, environmentally-friendly technologies and anything that's designed to go underwater. He lives in Edmonton, Alberta, where he spends a lot of time going over the handlebars of his mountain bike, hanging out in off-leash parks, and wishing the Pacific Ocean wasn't so far away.   All articles by Ben Coxworth

Oh yes. Because being hardware in a wifi network will totally make factory workers feel better about themselves.

Charles Gaines
5th January, 2012 @ 03:12 pm PST

I really hope that the intended use is to identify best practices and eliminate movements that could lead to injury.

The cynic in me sees new horizons in the field of micromanagement, though.

I do wonder if this could be used for general purpose motion capture, eliminating the need for cameras and ping pong balls taped to actors.

Jon A.
5th January, 2012 @ 03:16 pm PST

"Slaves equipped with sensors have been developed to time the actions of factory workers, in order to increase efficiency."

This is NAZIWARE to the max.

Mr Stiffy
5th January, 2012 @ 04:35 pm PST

I hope Jon A is correct and Mr Stiffy (does it not hurt being hard so long?) hope it is put to good use avoiding repetitive action injury rather than Mr Stiffy's POV

Bill Bennett
5th January, 2012 @ 09:16 pm PST

No sensor will ever eliminate the most wasteful of all production problems: dirt-stupid management.

William H Lanteigne
5th January, 2012 @ 10:30 pm PST

when will they realize that factory worker are not robot?

Cedric Boisdon
6th January, 2012 @ 05:50 am PST

This is great.

Workers that are slacking off or doing things incorrectly can be targeted for additional training.

Workers may not like it but really it is called WORK not HAPPY.

Keep your mind on the job , produce a good quality product and a the end of the day go home with no worries.

Captain Danger
6th January, 2012 @ 06:59 am PST

I'd be interested in whether this is a data gathering exercise for robotic arm manipulators. Could be a similar strategy to GoogleVoice being a proxy for speech algorithms. Either way, more safety and production control outweigh the big-brother aspect. Work is work.

Nick Beck
6th January, 2012 @ 08:25 am PST

Having run a stopwatch earlier in my career, I am keenly aware of the deficiencies involved with the process. The good thing about this product is that it is totally objective and thorough, with little opportunity of influence by either worker or supervisor. I like solid data. I think it will be too expensive and cumbersome as a micromanagement (dirt-stupid Nazi) tool.

Bruce H. Anderson
6th January, 2012 @ 08:45 am PST

This could be great for safety. You could program machinery to stop when the hand gets near a pinch point

6th January, 2012 @ 10:44 am PST

We are Borg, You will be assilimated !!!

and if i go for a waz, will they tell me how many shakes is optimum ?

Facebook User
6th January, 2012 @ 01:18 pm PST

1984 in the workplace.

Larry Hoffman
6th January, 2012 @ 01:51 pm PST

It's an updated version of the methods used in the early 20th century to study the motions required by various jobs and the time required to perform tasks. Some researchers attached small lightbulbs to test subjects' limbs and filmed them performing tasks in front of a grid.

This sensor sleeve combines previously separate techniques into a single instrument.

Gregg Eshelman
7th January, 2012 @ 02:05 am PST

Steve Rock;

No, there won't be a sleeve small enough to fit you down there.

Brian Hall
8th January, 2012 @ 07:02 am PST
Post a Comment

Login with your gizmag account:

Or Login with Facebook:

Related Articles
Looking for something? Search our 31,567 articles