Create a Steering Committee and Stay the Course

The first advice for sustaining POLCA in the long-term is to be resolute and disciplined about the system and the rules. Managers or planners are tempted to work around the system to meet some short-term requirements, but this usually results in a vicious cycle of breaking the rules and undermining the system. The resulting message to the shop floor personnel is that these rules don’t matter, and so eventually they start ignoring them as well, leading to a slippery slope ending in failure of the implementation. This is documented by the experience of Alexandria Industries (Chapter 10), where the authors clearly advise at the end of the chapter that you should “Stay tenacious and disciplined.” As stated in that case study, their company got extremely busy and ran out of capacity. In the authors’ words, “As the chaos grew, we found ourselves trying to out-think the system … We thought that we could better control the job flow manually. It is easy to predict what happened next. On-time delivery diminished and WIP increased … we were expediting more and moving WIP around because it was in the way of what we needed to work on next. Looking back, it is clear that reverting to processes that didn’t work previously is not the answer! Hence our advice is that you should stick with the POLCA system and rules regardless of business conditions.”

Therefore, as one of the first activities to help sustain POLCA, you should institute a process to review and handle unusual circumstances or extreme business conditions without changing the basic POLCA rules. An effective way of institutionalizing this process is to create a POLCA Steering Committee that includes the POLCA Champion and representatives from a few functional areas as well senior management. You could see this POLCA Steering Committee as a logical evolution of the POLCA Implementation Team. The latter was created to justify, design, and then implement your POLCA system. Now, the overall goal of this POLCA Steering Committee should be to support and sustain the POLCA implementation in the long term; dealing with urgent issues that arise can be among its shorter-term tasks. For continuity, and in consideration of their knowledge of POLCA, you could consider having several people from the POLCA Implementation Team continue to serve on the POLCA Steering Committee. However, you should also consider some new members based on the different role of this committee. For instance, some additions could be from senior management—people that have the standing to review and change policy decisions in areas impacted by POLCA.

The POLCA Champion should be empowered to call a meeting of this committee if an urgent situation arises, such as unexpectedly high demand, unusually short lead time requests by key customers, or unanticipated reduction in capacity due to machine or employee problems. It is important to note that you can in fact deal with most of these situations while still working within the POLCA system rules. Here are examples of techniques you can use to handle unusual demand or unexpected short-lead-time orders. You can:

• Add cards to some loops for a limited period, and then remove them at the end of that period.

• Increase the planned lead time for some routings. (See Chapter 5 for a discussion on the choice between this and the previous option.)

• Change the due dates of a few jobs. When the Authorization List is regenerated (daily, or at the start of each shift, or even more frequently) the Authorization Dates of all remaining steps for those jobs will be recalculated and these jobs will be expedited or held back based on the changes you made. Note that the Decision Time rules will still be followed by all the cell teams for those remaining operations.

• Add temporary capacity to some areas. Chapter 5 gives examples of how companies plan for reserve capacity that can be kicked in at short notice.

• Outsource some operations. For example, if you are experiencing unusually high requirements for turning of large-diameter parts and you only have a couple of Lathes that can do this, consider sending parts to a subcontractor for this operation (and possibly one or two preceding or following operations). For this outsourcing to be feasible in a short lead time, it requires that you already have a plan for this possibility ahead of time, along with pre-approved subcontractors for various types of operations.

The main point to note with all the above solutions is that there is no need to tamper with the POLCA system; the shop floor operators will keep working to the standard Decision Time rules. For example, if a customer needs a spare part urgently and this order needs to be expedited, while at the same time management knows that another order is just going into safety stock at a distribution center, then it can decide to allocate capacity to the urgent order first. This can be accomplished by adjusting the due dates for these jobs as explained above, and the people on the shop floor might not even know that this has been done—they will simply go by the latest Decision Time rules as the jobs flow through the shop floor.

To be able to use the above solutions effectively, the Steering Committee should conduct some up-front brainstorming about the possible techniques it might use, and then it should get pre-approval from management for some of the decisions that may have a cost impact, such as adding temporary capacity, or outsourcing an operation. This approval can be within pre-specified boundaries—for example, “You can deploy up to 20% overtime in Cells X and Y, for up to two weeks in a given quarter,” or “You can subcontract up to 20 hours of Lathe work in a month.”

The main aim of requiring the POLCA Champion to call a POLCA Steering Committee meeting is to ensure that these mechanisms aren’t being misused or overused. For example, the approach of changing due dates for jobs seems simple enough to implement, and managers might be tempted to do this regularly. This would just lead the way back to the pre-POLCA situation of constantly changing priorities. So, the committee could act as a check on such practices; simply requiring a committee meeting and then also needing the committee’s approval would help to filter out unneeded usage of this technique.

Over time, the POLCA Steering Committee can enhance the techniques it is using to deal with exceptional situations. It could learn how to better utilize some of the approaches, or come up with additional techniques based on suggestions from employees or committee members. This learning will also help with sustaining the POLCA implementation in the long term, an issue that is discussed at the end of this chapter.

The POLCA Steering Committee should also monitor the methods in place for training new employees and temporary workers on the basics of POLCA, and ensure that these methods are consistently used for such additions to the workforce. It should also engage in regular communications and updates for employees on the performance of the POLCA system. This leads us to the next section.

Track the Key Metrics and Celebrate Successes

You must keep tracking the key metrics that were set in place at the POLCA launch (see Chapter 7). It is best if some of the high-level metrics are published regularly for the whole factory to see, preferably in easy-to-read visual charts.

You should also continuously evaluate the fairness of the metrics. You don’t want to penalize teams for performance degradations that are due to events which are out of their span of control: Chapter 7 gave several examples of such situations. The employee feedback mechanisms discussed in the next section can assist you in monitoring the fairness of the current metrics.

Plan to recognize and celebrate when pre-specified targets are achieved. For instance, if you have historically struggled to get your on-time performance better than 70%, you could set targets of 80%, 90%, and 95% as goals and, as each is reached, arrange for some formal recognition from management along with a small celebration for employees. Don’t feel that these celebrations need to be large events. The literature shows that even small amounts of recognition go a long way toward making employees happier. For instance, a pizza party or cookout for employees and their families at a neighboring park gives enough of a message to everyone that management recognizes and appreciates what has been achieved. In addition, such events help to reenergize and spur people on to achieving the next goal.

Use Surveys and Feedback Forms to Assess the Qualitative Benefits

As has been stressed throughout this book, a good part of the success of POLCA comes from the fact that it is a people-oriented system. So, its success depends heavily on the buy-in from your workforce. Hence, you shouldn’t judge POLCA based only on “hard” performance data, you should also look for some “soft” data based on people’s opinions and impressions.

A survey form is an excellent instrument to gather such soft data. You should work with your Human Resources (HR) department to design such a survey, and ask employees to fill it in periodically, for example, one month after the start of POLCA, then three months later, and six months later. Typical questions that companies have included in their surveys are in three categories:

1. Employee understanding and effectiveness of training. Do employees understand why the company opted to implement POLCA? Do they feel that they understand all the workings of the POLCA system? Was the training adequate? Did they get sufficient opportunities to have their questions and concerns addressed?

2. System performance. In the employees’ eyes, have there been improvements in performance? For example, do they have less WIP and more space in their cells? Are there fewer hot jobs? Less frequent schedule changes?

3. Morale and satisfaction. Did employees feel they were sufficiently involved in the implementation process? Has POLCA improved the teamwork in their cell? Is the communication with upstream and downstream operations improved? Do they feel less stress in their jobs? Is their overall morale higher than before?

Even though we have used the term “soft” for such data, you can use a widely accepted method to turn the data into quantitative metrics that you can track over time. The method involves using an ordinal scale for the answers. For example, in response to a question about whether a particular problem is better or worse than before, employees could be asked to rate their answer from 1 to 5, as follows:

Alternatively, the question could be turned into a statement such as: “I have to deal with fewer schedule changes every day.” The employee could then be asked to select one of these 5 responses:

Both these approaches enable some statistical analysis since the data from surveying a number of employees can then be summarized using a histogram or median for each question. Note that for such ordinal answers, the mean is not a correct summary statistic, only the median or mode may be used; the median is arguably a better statistic than the mode, as it is based on the whole sample of answers. By looking at the histograms and medians for the answers over several time periods, you can see if you are making progress on a particular issue.

The final part of your questionnaire should allow employees to write open-ended answers. Typical questions companies have used for this portion include the following: How has POLCA affected your job on a daily basis? What improvements have you noticed in other areas outside of yours? Has POLCA affected how you deal with other areas? Do you have any suggestions for the POLCA Steering Committee on how to improve the system?

These answers should be collated in categories and reviewed by the POLCA Steering Committee. If HR is not represented on the POLCA Steering Committee then a representative from HR should also be involved in reviewing these comments as well as the statistics from the ordinal questions. Based on the collection of answers, the POLCA Steering Committee can decide if it needs to craft a response for employees and, if so, this can be included as part of the regular communications and updates by the committee that were mentioned earlier.

A last word of advice. It is best if you make the surveys anonymous—and make sure employees trust that they will indeed be kept anonymous! This will enable the truest data collection.

Conduct Regular POLCA Audits

To maintain the integrity of POLCA you need to plan audits of the system. You should consider two types of audits: (i) periodic audits, to check on the status of cards, and (ii) random audits, to ensure the POLCA rules are being followed.

Periodic Audits

Let’s begin by discussing the periodic audits. Such audits have several purposes: to examine whether the number of cards in a given loop needs to be reevaluated; to see if the procedures in a given loop need to be improved; and to check whether any cards are missing. This audit involves taking a snapshot of each POLCA loop in turn and recording the location of the cards in that loop. This data can be summarized for all the loops in a simple visual chart. Let’s go over an example to illustrate the details of a typical audit and the results.

Suppose a company has decided that, as part of the POLCA card handling procedures, the moving of a job and associated POLCA card from an originating cell to a destination cell will be done by material-handling operators, but the returning of a card from the destination cell to the originating cell will be everyone’s job, as explained in Chapter 5. In this situation, we decide to collect statistics on the location of each POLCA card based on the following seven possibilities (the abbreviations on the left side of each row correspond to the headings in Figure 9.1):

Orig. Board:	On originating cell’s POLCA Board
Orig. Cell:	In process in originating cell
Moving:	Job and card being moved to destination cell
Dest. Board:	On destination cell’s POLCA Board
Dest. Cell:	In process in destination cell
Returning:	Card being returned to originating cell
Missing:	Did not find card during this audit

Note that these categories are meant as examples; you can either combine or further divide categories to suit your situation. For example, if the moving of the job and card will be done by the team in the originating cell, you could include the “Moving” category in with the “Orig. Cell” since the move can be considered an extension of the process in the cell.

Figure 9.1Spreadsheet used to summarize data from a POLCA audit.

The audit team should then record the location of each card, by serial number, in a sheet similar to the one shown in Figure 9.1. The figure shows as examples, data for two loops: the A/B loop and the B/G loop. Entering the data into a spreadsheet, with a “1” for the location of each card, allows you to easily perform statistical analyses on the locations of cards as also shown in Figure 9.1 in the last two rows for each loop. An immediate observation from the “%” row for both loops is that, during this audit, around a third of the cards in each loop are in the process of being returned. Let’s say that during the POLCA card calculation you expected the return time to be relatively short so it did not contribute much to the number of cards in the loop, then this audit shows that in actuality there is a significant reduction in the total number of available cards for cells to use in production. This might just be a temporary phenomenon, so you could decide to monitor this situation over the next two weeks or so. If this continues to be the case, then it means you need to look into the process of returning cards and see what can be done to make this happen more quickly. The POLCA audit helps provide such clues that you can follow up on as needed—without the audit, this situation might not have been noticed for a while.

In some cases, you may need several audits to get data that is statistically valid. Again, having the audit numbers in a spreadsheet enables easy analysis over several samples. As you conduct more audits, you can view the tables side-by-side and also summarize the data using averages in order to detect any patterns. (Note that if you decide to go with an electronic POLCA system, such statistics can be continuously collected and easily summarized by the system without need for manual audits.) For example, if it seems that for a given loop cards are mostly in use and there are no free cards on the originating cell’s POLCA Board, you may consider increasing the number of cards. Conversely, if there are always several cards of a given type available on a POLCA Board, you can remove a card and tighten up the system.

The nature of the periodic audit is such that you do not need to stress about doing the whole audit at once and requiring a large team for this. You can audit one loop at a time and complete the audit over a few days. The aim is to get a snapshot of each loop and so any random moment in time is okay for such a snapshot. We emphasize the use of the word “random” here, because you don’t want to always use a fixed time like the start of a shift or the end of a break, because that might lead to some bias due to starting conditions.

Since POLCA cards have serial numbers, as part of this process the audit can help to determine if all cards are accounted for, or if a particular card is missing. If a card is missing, don’t replace it right away—our experience has been that missing cards have a way of turning up unexpectedly! You should wait for a period of time (such as a week or two) and check again to see if it has surfaced. In the meantime, if you feel that a particular loop is starving for cards—say it only had four cards to begin with, then a missing card would reduce the number of cards by 25%—you can deploy a Safety Card. The advantage of using a Safety Card is that it is always removed at the end of the loop, and requires an active decision in order to be used again. This will remind you to check for the missing card, and also force you to reexamine whether the extra card is still needed. If the missing card still has not appeared after a given period, such as two weeks, companies typically “retire” the serial number of that card and issue a card with a new serial number. This also ensures that if the card does show up at a future time, you don’t end up with two cards with the same serial number!

In terms of the frequency of the audits, you should start with more frequent audits and then lengthen the period between audits as your faith in the POLCA system increases. Initially, companies have used weekly audits until they felt they could move to monthly, and then quarterly. Since demand can change significantly over quarters, it is advisable to continue with conducting the audits at least quarterly.

Use POLCA to Drive Long-Term Improvements

Another mechanism that will help support the POLCA implementation is to make clear the connection between POLCA and various continuous improvement activities in your company. Use the data as well as anecdotal experiences generated through the POLCA operations to motivate and support improvement activities. As examples of this:

• The Safety Card mechanism used to deal with occasional component part shortages can also be used to gather data on the root causes for these shortages and their impact on the system (see Chapter 6). Analysis of these data can point to the most critical areas to tackle for reduction of the shortages.

• Management notices that a particular team often complains about waiting for cards from a downstream cell. However, according to the current data on workload, capacity, and planned lead times for the downstream cell, there should not be a problem with the number of cards available. This can result in an investigation of the data in the planning system as well as the operation of the cell: perhaps jobs are more complex than estimated; or there is more absenteeism in the cell than management was aware of; or a given machine breaks down more often than the planners knew; and so on. Whatever the root cause or causes, this will give some directions for improvement activities.

• Employees in a particular POLCA Chain that is currently using the quantum rules point out that if batch sizes can be reduced by around 50% they could use a single POLCA card to deal with each work order. This would simplify the operations, including the decision-making, material-handling, and control of an order that is split into multiple units. This provides a concrete target for improvement activities: what needs to be done in order to enable the smaller batch sizes to be used without overloading the system? For example, do some of the setups need to be reduced? Could some of the operations be done differently or performed on a different machine?

As such instances arise and improvements are made, the POLCA Champion or Steering Committee should be sure to document these occurrences as well as the results—such as quality improvements or shortage reductions—that ensue. The formal recognition and celebration events suggested in an earlier section will help to underscore these achievements. Over time, this will demonstrate to management as well the employees that POLCA is supporting long-term continuous improvement at your company. The message this conveys is that the fabric of the future of your company is interwoven with the successful operation of POLCA, and this is a powerful way to maintain support for and sustain the POLCA implementation.

This concludes the second part of this book, which has provided a systematic guide to implementing POLCA. Part III of the book serves to demonstrate that POLCA is not just a theoretical concept; it has been implemented in factories around the world with impressive results.