This is such a great question – posted recently by Grant Eldred in the AME LinkedIn Group http://www.linkedin.com/groupAnswers?viewQuestionAndAnswers=&discussionID=246005920&gid=730737– that I just had to respond! I’ve re-posted my reply here since it’s a common issue for many Lean practitioners…
Great question Grant!
How times change! When I first started out many manufacturers employed highly trained full-time Work Study / Industrial Engineers and developed actual or synthetic Standard Times down to the decimal minute. And let’s not forget that Taichi Ohno and many of his Toyota colleagues were also highly trained Industrial Engineers!
Very few organisations take that approach these days, and in my view have often gone too far away from facts, data and measures when it comes to labour times and costs. One of the (sensible) reasons for this is of course the fact that fewer and fewer manufacturers in developed economies actually manufacture high volumes of standard products where labour costs are critical and need to be measured very accurately. There are obvious exceptions of course, like automotive (Nissan’s UK plant here used to work to Takt times of around 1 minute).
BUT – and it’s a big BUT… Lean focuses mainly on all of that non-value-added time (often more than 90% of total lead-time) eaten up the Seven Wastes, rather than on the Value-Added activities that you’d be timing.
In practice I always involve the team, train them in basic Lean Awareness (20-30 mins on Lean and the Seven Wastes is OK), and help them come up with estimated times. I don’t aim to train them in Work Study techniques as well but I do always emphasise one of the basic points about arriving at a Standard Time – it’s the time that a competent trained operator can be expected to maintain over a full shift, day in, day out. So any observed times have to include a “rating” of the operator’s performance, and the observed times adjusted accordingly.
Before getting into the detal it’s usually possible to make a start by collecting some “gross / total” data on output rates and number of operators to arrive at some overall averages of labour content. Observing the work flow, looking for imbalances and estimating operator performances can then lead to some useful work with the team on line balancing, standard work, waste reduction, mistake-proofing, One Best Way and all of the other good Lean stuff (we’ve a blog post on the “people side of improvement” athttp://manufacturingtimes.co.uk/2011/07/09/increase-factory-output-part-3-targets-feedback-recognition-and-reward/ ).
Generally, this will lead to massive improvements in output and productivity, without any increase in worker effort, and for most instances that is all that the organisation wants / needs.
Occasionally though – and in my experience this is fairly rare – you can reach a point where you have developed a fairly “Lean” operation but it’s obvious that operator performance levels are pretty low (I’ve seen 20-50% levels). If at that point you’re in an environment where labour costs are highly significant then that’s when I do actually bring in the highly trained Work Study Engineers and develop more accurate Standard Times. This is an expensive exercise so often the objective will be to create a database of “synthetic” / parametric (“formula-based”) times that the client team can then manage themselves.
For what it’s worth – I’ve always employed (Industrial) Engineers and trained them in Lean, rather than the other way round. But as an Engineer myself maybe I’m a little biased!
Hope this helps!
As always, your own comments and feedback are greatly appreciated…