Archive for the ‘Nuggets’ Category
You are currently browsing the archives for the Nuggets category.
You are currently browsing the archives for the Nuggets category.
Sustaining Continuous Growth
It is common to see, in mature TOC production implementations, that once sales have increased and the CCRs were elevated, the number of red orders might start to increase. If no corrective actions are taken, further increase in sales will cause the DDP performance to deteriorate (number of blacks increase) which endangers the decisive competitive edge of the company. Since in a VV implementation we want to continuously increase sales, the S&T guides us to put in place the mechanisms that should be used for further increase of capacity and for adjusting the buffers. The intention of this nugget is to deepen our cause and effect understanding that underlie those mechanisms.
First, let’s remind ourselves the reason for the fact that in so many environments that claim to have no excess capacity, once DBR (and more so when BM) is implemented a considerable amount of capacity is revealed. The answer is that in the conventional mode of operation, which is dominated by the desire of people to look busy (and the need for high level of management interventions), the free capacity cannot be estimated by observing – or even measuring – the time that the relevant resources are standing idle. The only effective way to reveal the capacity is to change the mode of operation from pushing to be on the right hand side of the “U curve”, to operating within the plateau.
The TOC Insights into Operations will deepen the understanding of TOC beyond the knowledge available in any other material for Production. They were carefully produced to ease its usage and maximize the learning experience. The users can learn at their own speed and easily review any of the material, as they need.
The audit process is based on investigating “mysteries”; or as Eli Schragenheim had called them in his landmark presentation in Cambridge – surprises. A mystery is an effect in reality that contradicts a predicted effect of the S&T tree. In a recent audit of an MTO company (producing metal cables of the kind that is used for holding elevators), two mysteries were evident. But first a little background:
The company had already implemented, to the letter, the left part of the RRR tree and had quickly improved its DDP from around 50% (which is typical to its industry) to the satisfactory level of 98%. They proceeded by properly exploiting the resulting competitive edge of this extreme reliability, to the extent that the market is no longer the constraint. Actually they are winning so many orders that the “load control” pushed their promised due-dates to be later than the market lead time. Since it takes a long time to add capacity (about a year) they are now contemplating a mechanism that would enable them to safeguard their preferred clients (“When given delivery lead times are (much) longer than the industry standard lead time, not only may orders be lost, but clients may be lost.” – RRR S&T step 4.13.2 – Capacity Elevation).
(MTA combined with MTO)
In MTA environments there is a need to hold protective capacity. The problem is that in order to use the protective capacity when needed, this capacity is on average idle. In the previous nugget (“light blue 1″, no. 11) we discussed a way to directly exploit the protective capacity; to use this idle capacity to satisfy more sales. The first light blue method is based on the use of protective capacity to produce to stock, which is then offered in segmented markets – dumping markets. The nugget ended with a warning not to use this method if the company cannot find suitable dumping markets. In cases where the company can’t find enough such markets, and consequently a part of the protective capacity is left unutilized, there is another possible method that can enable exploitation of the remaining free protective capacity. This method is based on processing clients’ orders by the protective capacity and therefore it is applicable only in mixed MTA+MTO environments.
One of the crucial elements required in any MTA environment is maintaining high enough a level of protective capacity1. Practically, it means that the work centers are not allowed to operate 100% of the time. In fact, the most loaded resource will be, on average, about 20% idle2. However, as we all know, people feel very uncomfortable with the idea of idle capacity. This is especially true in environments that utilize highly expensive equipment, and even more prominent in environments where full activation of resources is theoretically possible, since they have saleable products at many different intermediate stages of the production process. Such environments do exist, and are evident in many V environments.
The reluctance to hold capacity idle may cause companies to dip into the protective capacity, especially when we bear in mind that most companies are not used to operating with a decisive competitive edge and therefore, at least in the early stages of implementation, are not fully aware that having a decisive competitive edge does not only help in getting more sales, but is the foundation for stable growth. How can we make sure that the proper amount of protective capacity will be maintained?
In many companies, some SKUs are consumed on a regular basis, whether by one or multiple clients, while other SKUs have a sporadic consumption pattern, either because the demand is rare, or because the SKU is customized for a particular order of the client. In these environments, regular consumption should be supplied immediately from stock (MTA), while sporadically consumed SKUs (as well as exceptionally big orders of regular SKUs1) should be made only to clients’ orders (MTO); in other words, they should operate as a partly MTA, partly MTO environment.
The mechanisms to control both the MTA and the MTO in a mixed environment are the same mechanisms used in an environment of pure MTA or pure MTO. The only difference is that when using load control for providing promised due-dates and for determining the time to release the material for MTO, we regard only the MTO production segment instead of taking into consideration the total capacity and load. More explicitly, the way to allocate the capacity in such mixed environments is as follows: first assign capacity for MTA based on the current coverage; then add 20% protective capacity (Note: The needed protective capacity is not 20% of the total capacity, but only 20% of the capacity required by MTA). The capacity remains is allocated for MTO.
A company that had started Viable Vision implementation about a year ago had made satisfactory progress. Nevertheless, there was a point that troubled them as well as us: the level of inventory in their regional warehouses did not significantly decrease. It was months after the establishment of a central warehouse, coupled with the activation of daily replenishment according to actual consumption throughout the supply chain. Thanks to having a good level of availability at the central warehouse (99%), the replenishment time to the regional warehouses was cut to a mere fraction of what it originally was. Setting the inventory targets in the regional warehouse in accordance with the shortened replenishment time should have reduced the original high inventory levels there to less than half, even when taking into consideration the additional inventory that was needed for raising the DDP of the regional warehouses from the original performance – below 50% – to the current delightful level of 99%. More than enough time had passed to enable the mountains of excess inventory to be flushed out. So, how come the inventories in the regional warehouses were lowered by just 10%?… Click here to continue reading.
When we are told that a production company has 20% red orders, what does that mean? Seemingly the answer is clear – it means that if we go to the floor, we will see that 20% of the orders are red. But then again, maybe it means something else – that out of all the completed orders, 20% were finished while being red?
Here we have a situation where one term is being used to describe two different things. This ambiguity wouldn’t be a real problem if the practical implications, stemming from the two different interpretations, where the same. But this is definitely not the case; one of the above meanings of “red” is very important in terms of assessing the state of the operation, while the other has no practical implications and using it leads only to confusion.
In order to clarify the above statement, we ought to better understand the meaning of each definition. The first definition of red percentage is a description of a snapshot – the situation on the floor at a certain moment: how many orders from each color do the production workers see? Figure 1.a schematically shows a distribution of orders’ color on the floor at a specific point of time, under the assumption of a constant stream of incoming orders. The X axis is not a time axis, but rather represents the “age” of an order at the point of time when the snapshot was taken… Click here to continue reading.
There are times that a technical consideration at a certain work-center forces us to process a minimum batch which is bigger than the order (e.g. a mixer that requires a minimum of 20 liters). Then, of course, we have no choice but to work on more units than needed for the order. But that doesn’t mean we have to carry those extra units onwards through all the processing chain (after passing the work center responsible for the technical consideration), as is too often done… Click here to continue reading.
The effectiveness of the priority system stems from the fact that it is robust yet simple; only three priority colors, with a strict instruction not to try being super- accurate and pin point which among the same-colored orders should be processed first. That being said, there is one exception – a unique case where refinement of the priority system is very much needed.
The case we refer to relates to situations where there are product families with considerably different production buffers. In such a situation, a product with a relatively long production buffer might have, at some work-centers, touch time which is indeed negligible relative to its own production buffer (< 10%), but is significant in comparison to another, shorter production buffer (> 1/6). Figure 1 schematically demonstrates the described situation… Click here to continue reading.