Wednesday, 5 May 2010

LEAN and mean

Had an interesting call with a Gartner analyst this morning about LEAN - an on-going process improvement technique, originally developed for the engineering and manufacturing industries but now applied to any process. It focuses on a number of basically common sense approaches (but we all know that common sense isn't always that common), and I like it because its simple and makes sense.

A key principle is that of removing waste. Waste can be a number of things. Stuff for example - detailed project plans that sit on a shelf and are never referred to, piles of paper people print and take to meetings, documents that are produced too soon and in too much detail. Papers should be read, not weighed. It can also be unnecessary steps in a process such as a sign-off step just to keep someone happy, or unnecessary loops.

Another waste can be motion - waiting for information, not knowing where to get help and information from, unnecessary movement of information, people or things.

Time is another form of waste - task switching is a good example. If you're in the middle of doing something and someone phones you up with an unrelated 5 minute question, it can take 30 minutes out of your day.

In IT terms, over-development is a waste - good enough is exactly that. We should be aiming to deliver just what is needed which might only be 60/70% of what was asked for, and delivered quickly.

The LEAN technique involves following a process through from end to end and working out what is waste and what adds value - if anything isn't adding value it should be remove. this technique brings together staff from different functional areas and allows them to see the whole process rather than just their bit of it, and should help to improve collaboration and reduce silo working. By allowing staff to work in teams on processes (without necessarily having the line management present it should also empower them to suggest improvements. Of course, with all such techniques it's important that it is kept lightweight, that quick wins are achieved and it doesn't become an end in itself. I'll certainly be suggesting that we give it a try some of our processes - the difficulty will be deciding which ones to start with!

2 comments:

George Credland said...

More than happy to volunteer some processes...

I'm currently waiting for details of requested improvements to myJob (Employee Self Service) and myTeam (Manager Self Service). Aims include extending the scope of what people can see and more significantly maintain for themselves without central intervention.

We already apply a number of the techniques from LEAN development:
http://en.wikipedia.org/wiki/Lean_software_development

e.g.
User feedback sessions are held for significant new developments. With both eRecruitment and Fixed-Term contract changes iterative changes have been made to address issues identified.

Developers are now able to attend User Groups sessions to hear issues directly and have a better understanding of what's going on.

I disagree with the build it first approach which would waste a lot of development time. Instead ideas for screens/web pages can be sketched/prototyped as part of detailed design discussions. This needs to come after the overall aims and flow of information are resolved otherwise time is wasted on cosmetic layout issues rather then more significant matters such as where/how the information feeds in and how its handled through to the final result. e.g. quickly building a screen for person A to enter information collected from person B, when access could have been given to person B, removing the delay and re-keying errors in passing the data to the second person.

Of course this can only work with buy-in from other departments as we don't "own" most of the University processes.

Pablo said...

Good stuff, and our work with Student Services (with advice from colleagues at Hallam) on developing a straightforward method for analysing and improving processes will hopefully contribute.

'Lean' has sometimes though got a bad name for looking at 'value' from an administrative or narrow cost point of view rather than genuinely reducing waste and maximising benefit for the service users. We'll need to take care it is used to do the right things and do them righter, not to do the wrong things righter.