A Guide to Implementing the Theory of
This Is A Local Labor Intensive Manufacturer with a solution right out of The Goal. The firm had undergone very significant growth both domestically and in a substantial export market but was constrained and couldn’t meet current orders. In fact the theme was; “you don’t understand, we have no additional capacity!”
The initial constraint was in the polish shop. Using very limited training materials – just the dice game and The Goal – and some improvements in operations, we were able to increase output by some 25% and reduce the work-in-process by half in just 12 weeks on constant head-count.
Of course the constraint moved. In fact, it moved right around the factory in several stages, until it came back to the step before the polish shop, which was assembly, and that is where we choose to keep it from then on.
Output continued to rise to meet the market – a 42% increase by the 8th month – all on constant head-count. Many types of product were reduced in batch-size by a ½ to a ¼ to an ⅛. Frequency was increased proportionally. Stock-turns for major lines grew to several times per week. When you can ship next day ex-stock your market grows faster than your competitors as retailers seek to replace other companies’ goods that they have sold but can’t yet replace. Replacing other firms’ floor stock is a decisive competitive advantage.
Let’s Restate This. Because many people miss the point – or never see it – a 42% increase in output = a 42% increase in sales. This is because we only replaced what was sold and we could sell much more. But how much of this is additional profit? Well the answer is about 66% of it. The remainder, 33%, went to pay for raw materials and other 1:1 variable costs.
66% of a 42% increase in sales = 28% of initial turnover is now additional clear operating profit. Slot in your own turnover and ask yourself whether 28% of that as additional operating profit would be nice or not?
Initial operating profit was about 33% of turnover. So another 28% equates to something like; (33+28)/33 = 1.85 – almost double – the original operating profit. That is a very sound and sustainable position to be in, in a very short period of time.
Assign let’s say a generous initial net profit of 10% and we have (10+28)/10 = 3.8 – almost quadruple – the initial state.
This is not unusual. This is what Theory of Constraints is so good at. Undoubtedly most of the new cash is reinvested into the business, and why wouldn’t you do that? You can retire old debt faster; you can make new capital expenditures, you can make new discretionary expenditures in marketing, development, and so-forth that were previously beyond reach. You can operate with a great deal more confidence.
Remember, this is not a one-off, such as clearing out trapped work-in-process; this is a continuous and sustained improvement in productive output and cash inflow, not for the duration of the improvement, but for the next year, and the year after that, and the year after that too. It has important ramifications not just for cashflow but also the overall valuation of the business. People who own or invest in such businesses will be most mindful of this.
Such an increase, with the exact same people, with the exact same equipment, is the consequence of applying some significant and codified know-how rather than relying on old accepted practice.
A Different Scale And A Different Country. First a personal note. I lived and worked in Japan for a total of 4 and a half years, off and on. Three of those were; first in the city of Okayama at the Hiruzen Research Institute, Okayama University of Science, and second in Tsukuba Science City at the Geological Survey of Japan while doing post-doctoral research. These were, in a way, cultural enablers to what follows.
And what follows is in the heartland of Toyota, Aichi Prefecture, near Nagoya, and indeed at a major supplier of Toyota and Honda. The quality of the output is world-class, but the firm had still consistently “declined” the invitations of Toyota to “help” over its production planning for many years. This company is – as are the vast majority in Japan – a batch producer. And batch producers know how their operations are fundamentally different from the final stage assembly line for which Toyota is justifiably world-famous. These facilities had reached the point where they felt that they had no additional capacity to meet rising demand.
In this particular plant, we were able to use the production simulator with all management from the factory manager to foreman. Having a professional planning function also provided a challenge at first and a substantial benefit later on.
We increased output by 18% in 12 weeks on the major throughput lines in this very substantial operation. In addition we were also able to very significantly reduce both the work-in-process and the lead-time. When you focus on the constraint, you can improve the whole plant very quickly.
This photo was originally used in a corporation-wide meeting day of plant foremen and was entitled “No Red Lights.” This particular machining stage had been identified as the constraint to overall production and the greatest single challenge was to synchronize setups in order to produce a smooth flow. One set-up operator cannot be in more than one place at one time as you know, and the other places while waiting lose significant production, therefore to plan a set-up sequence for one place at one time has tremendous benefits in overall output for the whole plant. It would not seem believable given the number of overall stages in the process, but remember everywhere else has additional capacity. It a salutary lesson for people that a small change in a small part of the plant can have such tremendous benefit overall.
And because this aspect of constraint planning is so important and so underrated overall, here is a photo of the planning boards for the constraint machines that we saw above.
Although this board is in its “infancy” here, it later became significantly more visual with photographs of the individual set-up personnel, as well as the machine and job and time as shown here.
The professional planner is taking, the local information from his master schedule and making it visible. The constraint is the only place where we had such a time and sequence schedule. The remainder of the plant operated on fist-in-first-out. Always make things visual.
Although batch sizes had been considerably reduced over historic values, we were still able to increase output with increased set-ups on the constraining resource. The foreman also “liberated” one additional constraint machine by taking it out of a robot line and operating it manually. People are very resourceful when they have focus.
Now, of course, this would not happen in your company, but some set-ups required a laptop to talk to the CNC machine tools, however, the PC would often be somewhere else, and not available. That did change too. Constraints bring about unbelievable focus to the situation.
In another part of this same plant we had a small sub-plant which dealt with micro tool-bits.
This is just to reinforce that automation is not always automatic.
If you look closely at the andon lights you will see that the machines are almost in unison “orange” and all are requiring attention of some sort at the same time.
Despite the sophistication of the machines, and the phenomenal accuracy of the whole process, there were intractable problems in the planning of this particular work flow.
Even when the work-in-process is physically minute, you need to know what it represents in terms of time. Enough said.
This Is A Different Plant In The Same Corporation. I climbed an abrasion cleaning tower to get this shot. You can see a some of the work-in-process in the foreground of this shot. There was literally “weeks” worth of work here in the initial stages.
In this particular plant we once again took the full management through the production simulators. The planning manager also took to using the Thinking Process tools with some proficiency. Personally I probably drew more clouds on the whiteboards within this plant than I have anywhere else – but that is what clouds are for, they are tools of immediate engagement for use on the fly and for use on the floor.
In this plant we achieved what very few batch producers in Japan have been able to do, we reduced a very significant work-in-process back-log by ⅔ in just on 12 weeks.
This work was well reported in the national Japanese business press and also national television. It was a phenomenal result when you understand the challenge to people’s sense of security in seeing the “mountains” of comfortable work disappear. It is also a tribute to the plant manager and all of the staff that we could achieve this. Work on the floor decrease from some 600,000 pieces to just on 200,000 pieces over this period.
Now, here is a reality check. This is not one of the constraint areas; this is part of a non-constrained line – and that is most of the plant after all. I wanted to show it to you to reinforce that there are people everywhere. Also the machines here are much older and much more manual than you might expect, set-ups are therefore much more intensive on these types of lines.
If ever, you are going to get a flow in a plant such as this, you must decrease the size of your process batches. Generally, this is not somewhere where you can make use of smaller transfer batches, there are things that “bulk” them up again, for example furnace time during heat treat. So smaller process batches are the only way to get better flow.
Smaller process batches, as any operator will tell you, carry a penalty, they incur more setups. In all non-constrained areas, however, there will be considerable time available; the trick is to keep building that spare time into the system.
Building spare time into the system, ahead of the need, is part of the continued subordination of the non-constraints to the constraint. It is part of the whole process and it has to be actively managed. Too often people fail to understand this. These areas are prime candidates for kaizen activities.
Constraint Schedule Boards. One last photograph from this plant. The plant basically had 4 lines running within it, based mostly on the physical size of the work being produced; small, medium, large, etc., That means that we actively managed 4 constraint areas, one for each line. This “spot” is in front of the foreman’s station of one such area. It contains two boards that have the time and sequence for each machine and for all of the immediate work in the constraint area.
Here we can see the plant manager (arms folded) as the section manager for the area and the planner for the line discuss work for the area. The other participants are the supervisors of all of the other sections of the plant.
This particular board took on a “visual life” of its own with many personalizations – photos, scripts etc. Most importantly it recorded when we expected to start and finish on a particular machine in that section and when we actually started and finished.
In all of the non-constrained areas, there were also work boards – but they contained only the sequence of the work based upon first-in-first-out principles. I cannot reiterate too much the benefit of going to the actual place, talking to the actual people and seeing the actual problem. And having been seen to have done so. Let’s change countries, scale, and type of business once again.
Critical Chain Project Management. Manufacturing has “short touch time,” engineering projects have “long touch time.” Nevertheless many of the fundamentals are the same, usually going under different names. Multi-tasking is just the same as having multiple batches on the floor in manufacturing and the consequence is the same.
This is a small light engineering workshop that designed and fabricated made-to-order road trailers for trucks (U.S. tractor). They also ran a repair shop and the two interacted in rather unhelpful ways.
Previous purchasers of trailer units who had mishaps that took the trailer off the road, and thus were non-revenue earning, and expected the shop to do the repair with utmost haste. The new trailers in production therefore slipped back.
Make-to-order and already long lead times encouraged purchasers to request changes to detail as the job proceeded. Again the new trailers in production slipped back.
The net effect was that this business was losing money. And while creditors will only ever foreclose as a matter of last resort, that is what happened here. We arrived one day, the receivers were there, there was nothing more that we could do.
The real value was in fabrication, the urgency was in repair, but the shop downtime in repair was never actively managed for via buffering. As I was once told, and have never forgotten; schedule according to reality.
Throughput Accounting/Receivership. Standard cost accounting assumes independence, throughput accounting assumes interdependence; they are profoundly different. Professional manufacturing cost accountants know this and are pragmatists at heart. Many others do not and therein lies the problem.
You can use throughput accounting in a interdependent process to drive profitability that standard cost accounting methods cannot replicate. Conversely and sadly you can use standard cost accounting methods in interdependent process to drive non-profitability and not even know it.
An example is a firm not too dissimilar to the one at the top of the page, albeit a little smaller, a little less professional. They had made a decision to change some of their sales “lines” and were building stock for this. The sales guy – probably as a consequence of his own reductions in bonus – saw some danger signals. I worked through 6 months of accounts with the owner and showed him that he was failing to generate operating profit. I offered to go with him to see his banker (banks understand cashflow) but he would not hear of it, his accounts guy had been with him for years. Sometime later I learnt that they had gone into receivership. Sometimes there is nothing more that you can do.
Sawmilling is one of those industries that – unlike manufacturing – cannot know the grade or volume of their output until after they have milled, or mined, or harvested, or processed, or whatever. This was a large regional pine mill taking saw logs from plantation forests.
The mode of operation in these mills is cut-to-grade, that is the head-rig operator makes decisions to optimize the recovery of the best grade of timber from each saw log. What you can see here however is a colleague with printouts of end-stage stock buffers, many of which are “in the red,” or worse – empty. We changed the operation from cut-to-grade to cut-to-replace using replenishment buffers that took into account the often considerable time required for various post-milling machining, treating, and drying operations.
The mill at the time that we were involved had significant cash flow issues. If you are cutting for grade recovery, you can have the best grade recovery in the world and if your customers are not buying that grade then basically you have sunk a huge amount of cash in a pile of timber that you cannot sell. Moreover you will not be cutting grades that customers are buying or indeed are waiting to buy. The buffers, rather than a forecast, tell you exactly what is going on and what grade needs to be cut now in order to replenish the supply in good time. Few people understand this. Sometimes the most expensive products will yield less cash for the time and effort involved than cost accounting would suggest. In essence you are not adding value you are simply adding cost. Again very important when you are cash constrained.
This is just one place where counter-intuitive and throughput-based decisions increases cashflow into the business. Part of the personal gratification was that we got to put in place a workable solution in an industry that we had always been told “we couldn’t understand.”
Problem Solving For A Central Government Agency. Something completely different; an agency of a government social service department used Bill Dettmer’s Strategic Navigation to produce post-it notes of symptomatic issues going up 3 walls of one of their offices. I was asked to bring some structure to the issue.
Although I prefer to employ the Thinking Process as a tactical toolset, this was more of a facilitation of a broader strategic issue.
The agency felt that it had an opportunity to move matters forward for the people whom it served. The services that these people received were fragmented across a number of service providers both within government and within the wider community.
From the large number of symptoms several key “threads” emerged which were developed into individual current reality trees. All of these originated from a single common dilemma. That dilemma was the core conflict of the system.
The end result, of which you can see just a part, was used to communicate the dilemma and its consequences to the Minister concerned.
Healthcare, a large tertiary regional hospital; 300 senior medical officers, 500 beds, 20 acres of floor space, and 80,000 in-patients per annum.
There are not many people who can claim to have systematically brought forward people’s birth times, but that is what we achieved here in obstetrics and gynecology for planned routine cesarean deliveries.
In the emergency department we introduced a 3 equal-zoned buffer board to implementation a national 6 hour target for assessment & admission or treatment & discharge. We also introduced a number of counter-intuitive measures to focus staff attention on improving patient flow through the department.
Also involved in acute & elective theatre productivity and start-times, sterile supply, hospital replenishment supply chain, integrated operations, and numerous smaller interactions. The hospital knew how much a 22 cent blood tube cost us, but not how much a 45 minute late-start in a scheduled theatre might cost. In fact it still doesn’t know. Theatres are a “natural” for bottlenecks, and the number of late-starts and incomplete kits was, and remains, chronic. Essentially these are issues of poor or non-existent subordination, it is not that the capacity doesn’t exist, it does, but it is wasted. Likewise, getting people into hospital is relatively well co-ordinated; getting them out again is not. Enough said.
Throughout this health system, not just in this hospital, acuity is used as a proxy for urgency. In all of the diagnostic suites, interventional suites, and also elective surgery, a static “schedule” was regularly updated (every several hours, or daily, or weekly as the case may be) for new cases of higher acuity. The consequence is that even as less acute cases matured dynamically, they were still pushed down the queue rather than remaining near the top. If you like the buffer status of the patient has no relevance when it should be of the most relevance. The discipline to do what is most important is always subjugated to what is deemed to be the most urgent.
Despite the apparent complexity, a large hospital such as this well knows from history the likely demand for each hour of the day, day of the week, and week of the year. The trick is to know where to focus and how to subordinate. Healthcare is still learning. As in other types of services there is abundant capacity but more often than not that capacity is destroyed. A profound lack of understanding of interdependency and of variability conspires to convince the participants that indeed there is no spare capacity.
Five acres of world-class Japanese factory filled to the gills with work-in-process and unable to keep up with demand might look like an intractable problem, it is not.
One of the largest emergency departments in the country with patients waiting on beds down the corridors might seem like an intractable problem, it is not.
A sawmill in a recession with saw-logs being driven past the gate for export overseas might look like an intractable problem, it is not.
The way we get ourselves into these problems is because of our way of thinking about them, and the way that we get ourselves out of these problems is a different way of thinking about them. And with that comes a different sort of skill-set and I hope that these stories will help you to understand that.
You know how to find me – the contacts page.
This Webpage Copyright © 2013 by Dr K. J. Youngman