Mura → Muri → Muda

Sent to my mailing list on 26 April 2019

"I didn't want to do this, but to make the numbers for our investors we're all going to have to give extra for the next few months. I'd really appreciate all your commitment and hard work on getting us over the line."

Ugh. Crunch Time.

We've all been there. The bosses gather everyone together and tell the team (without really saying so) that they expect us to work long hours to hit a deadline or the targets set by investors.

Early in my career I used to accept crunch time as part-and-parcel of working in tech. A young urban male with no responsibilities can easily accommodate short bursts of long work hours. But for everyone else, and for sustained periods, it's just not a winning tactic. For anyone involved.

In a 2006 post, lean manufacturing guru Jim Womack neatly describes why.

The Toyota Production System defines three areas of waste:

• Muda - waste. Any activity that consumes resources without directly creating value for the customer. Muda is further distinguished by necessary activities that can't easily be removed and those that can be addressed through kaizen (continuous improvement), of which there are seven types.
• Mura - unevenness in operation. Jim Womack gives an example of increases in automobile production that are not related to any increase in customer demand, but more about 'making the numbers' for the quarter.
• Muri - overburdening equipment or operators. In lean manufacturing, the consequences of running equipment harder or longer than their designs allow are well known. Similarly, overburdening people with long hours or unrealistic expectations of speed holds consequences that are also a source of waste in the system.

Womack's argument is that lean manufacturing practitioners tend to focus their initial improvement efforts on eliminating the seven types of Muda:

1. Transportation (inefficient movement of products and supplies)
2. Inventory (Storing too many or too few products or supplies)
3. Motion (Unneeded effort by workers or machines)
4. Waiting (for upstream work to be completed)
5. Over-processing (putting more work in than necessary to meet expectations)
6. Overproduction (creating too many products or adding characteristics that aren't valued)
7. Defects (issues that cause products to fall short of expectations)

You should be able to see how each of these apply just as much to software development as they do to manufacturing.

But Womack argues that many waste problems are a result of problems of Mura (unevenness) and Muri (overburdening), and so it's better to focus improvement efforts there first.

Crunch Time is an example of this.

Crunch time usually arises because there is a perceived need to deliver a large chunk of value in one go. There's usually a deadline to go with it. This is an example of Mura - an unevenness in the delivery of value over time.

Crunching in order to meet these expectations is Muri. It's an overburdening of workers and processes. Too often, the expectation is that it's possible to just turn up a speed dial with no consequences.

But in reality, overburdening leads workers to cut corners, force value through the system too quickly, and make mistakes. In other words, Muda. Take another look at the seven types of Muda above and mentally tick off the ones that you've made in the past during crunch time. I counted four. Counter-intuitively, it's even possible to over-process and over-produce in crunch time, even though we would expect the opposite to happen. Can you see why?

Unfortunately, as Womack states, fixing this "will be hard work and will require courage because it will often require you to rethink longstanding sales, management, and accounting practices that create the Mura and Muri."

In other words, it's essentially impossible to fix once crunch time is already on you. It's something that needs to be anticipated in advance, and the fixes are mostly cultural.

All the best,

-- Jim