Ian Glenday’s guide to starting levelled production
FEATURE – In the latest article of his series for Planet Lean, Ian Glenday explains the practical steps an organization can take to properly introduce levelled production in its operations.
Words: Ian Glenday, lean coach and author
My previous article for Planet Lean described how the first three steps of levelled production – or, as Toyota sometimes called it, patterned production – create "economies of repetition" that lead to two positive outcomes: faster learning curves and routines plus stability in planning and production. These, of course, boost performance and encourage the creation of an environment that supports sustainable continuous improvement as well as an easier application of lean techniques.
A fixed repeating pattern in production that involves more frequent short runs often seems impossible to achieve to many people. So let me tell you how you can do it.
Let's start from the quote from John Shook I shared last month. In it, one of the things he said was: "A pattern of running each part number repetitively, perhaps using an 80/20 rule to differentiate high from low runners, is established."
The 80/20 rule, also known as Pareto analysis after the person who first described this statistical quirk, is a clue about how to set off: start with the 20% biggest products that account for 80% of the volume. In the past I have used Pareto analysis extensively, so much so that I began to realize there were other statistical quirks in the numbers – and that these were far more significant than 80/20!
The first was that invariably just 6% of the products accounted for 50% of the volume of sales. It wasn't just products that this applied to. Here are some other examples: 50% of global military spend is by just 6% of the world countries. Just 6% of SAP500 companies make 50% of USA profit. Every time I have done the analysis for patients operated on in hospitals just 6% of all the types of operating procedures carried out accounted for 50% of the patients seen.
The new rule is 50/6, with 6% being the number of items to be put into a fixed repeating cycle and 50% being the volume being produced this way.
There are two other significant statistical quirks in the numbers. These are that 95% of the volume comes from 50% of the products and that 99% comes from 70% of the products. This means a staggering 30% of the product range typically accounts for only the last 1% of sales.
These repeated occurrences amazed me and I discussed them with friends and colleagues, to get them to run their own analyses to see what figures they got. Time and again, the answers came back: 50/6; 95/50 and 99/70. One friend, Carla Geddes, suggested a couple of ideas to make the analysis into a more recognizable analytical tool.
Her first idea was to give each category a colour code, to make it easy to remember and create a "label" for each group of products. Next, she said that because the 80/20 rule had been named after the person who first identified this quirk, Vilfredo Pareto, perhaps I should name my analysis following the same idea. She suggested I call it "the Glenday Sieve."
It is not too difficult to develop a fixed cycle for just 6% of the products. You can then value stream map the "green" items to help unravel the "spaghetti pathways" one usually finds. The result is a green stream for these few high-volume products with shorter throughput times, better flows, less activities due to routines, and much lower inventories possible for the green SKUs and associated materials.
The yellow SKUs are where you want to direct your capability improvement efforts. Change over reduction exercises and smaller batch sizes make it easier to introduce these products into the fixed green cycle. This typically results in a staggering 95% of the total volume, yet only 50% of the products range now running in a fixed sequence "every product every cycle" schedule.
The blues are a special category (there is no number associated with them). To many people lean is primarily about eliminating waste, yet a big source of waste – unnecessary complexity – seems to be rarely identified and tackled: the blue category is intended to bring this waste to people's attention. Unnecessary complexity adds cost but no value – I am talking aboutthings costumers either do not recognize or do not care about. For instance, what opportunities are there to harmonize the raw materials and packaging so that the final product appears different to customers but makes these SKUs easier to include in the cycle? Blues with actual examples from companies and what to do about them will be the subject of a future piece in this series.
The red SKUs will need to be carefully reviewed if we want to determine the real cost they represent for the business. The impact of these products to the total supply chain and overall company costs, including overheads, must be understood to be certain that the benefits of selling them genuinely outweigh the costs. It is not uncommon for companies to recognize through this analysis that they actually have two distinct businesses. One is high-volume products (the greens) for which the plant, business processes, and performance measurement systems were designed. The other is a low-volume operation (the reds) that is often more difficult to plan and carry out. The same equipment, processes, systems, and KPIs are being used for both green and red SKUs, to the detriment of customer responsiveness, efficiencies, inventory levels, and profit margins. Is it really a sensible business practice to continue running small volumes of red items on equipment that was originally designed for high volumes of green SKUs?
"Red products" are hardly conducive to the creation of lean enterprise. Let's look at Toyota for a moment: what they do is grouping their options into specific offerings based around the engine size, power and level of model trim in order to reduce the total number of variants they manufacture. As a consequence, they produce significantly fewer different SKUs compared to most of their competitors.
For some people having a fixed sequenced and volume cycle for just 6% of their products still seems impossible due to demand variability. The question you want to think about is this: how can you have a fixed stable plan at the same time as having variability in demand? Come back next month and you'll have your answer.