Motivation for Rethinking the Shop Floor System

For the first few years of its operation, Provan had used its ERP system together with Lean techniques like Kanban to manage its order flow. However, management noticed that Provan’s customers were getting more demanding: the variation of parts ordered had grown, and batch sizes in the orders had shrunk considerably. At the same time, customers wanted more flexibility and shorter delivery times. As a result, it became more and more difficult to satisfy customers using the material management systems that were in use.

The issues with the existing systems became more pronounced during 2012 in the production of stoves for a major customer. The stoves were comprised of 130 parts that were produced in several steps. These steps mainly involved laser cutting, bending, and threading. Then these 130 parts were welded or fastened together and assembled into one stove.

The parts were stored between the successive fabrication and assembly steps in a warehouse that occupied over 600 square meters (around 6,500 square feet). The stoves were produced in batches of 60, and the lead time from the initial laser cutting operation until the final inspection took about four weeks.

The customer was pleased with Provan’s performance, and asked the company if it could supply three variations of the stove product. This meant that three types of stoves needed to be welded and assembled, and 130 components per stove—often with different routings—needed to be manufactured. So, while management was pleased with the increase in the customer’s orders, it also feared that this would almost triple the needed warehouse space to store the components. In addition, they were concerned that the increased complexity could lengthen their lead times, which would also add to the WIP and storage space for semi-finished parts. For a small company, this extra space would add considerable expense.

Implementing the Stove Cell

As luck would have it, at about this time Provan was introduced to Quick Response Manufacturing (QRM) at a workshop given by Rajan Suri in Belgium. Management felt that the QRM approach could help attenuate the warehousing problem and the company decided to create a QRM Cell for the production of stoves.

In creating such a stove cell, a major issue that needed to be resolved was the organization of the complex work flow within the cell. Remember, three types of stoves needed to be fabricated and 130 components per type needed to be produced, often with different routings within the cell. How could the operators manage these material flows without needing to resort to a miniature version of an MRP system within the cell, with all its associated complexity and problems?

Introducing Color-Coded Visual Work Management

After six months of discussions and tinkering with ideas, an elegant solution was developed that would be simple and yet avoid the typical problems. This solution got the employees to become familiar with visual management and using color-coding to signal tasks and priorities, and prepared them for the companywide POLCA implementation later.

The solution involved the formation of a cell that would undertake all the processing steps except for laser cutting. Laser cutting was kept separate from the cell, as this expensive machine was shared among many different products and clients. Buying an additional laser cutting machine for the cell couldn’t be reasonably justified and wasn’t necessary to accomplish the goals. Other workstations such as bending, rolling, threading, grinding, and welding were moved and positioned in a U-shaped cell (Figure 13.1). A color code was marked on every workstation in the cell to support the visualization of the workflow. Once the cell was in place, the order steering system, described next, was used to manage the flow of work.

Figure 13.1Stove cell. Note the material on carts in the middle and work stations on the outside of the cell.

When parts for a batch of stoves arrive from the laser station, they are put on a set of carts in a standardized way as follows. First, a numbered metal flag with a specific color is attached to an empty cart. The color on the numbered flag corresponds to the stove type. So, with three types of stoves, three different colors are used. The number on the flag indicates which parts should be put on which cart. Multiple parts are grouped together on a cart based on their routing within the cell. Then on every cart, a row of additional colored flags is added. These flags indicate the routing the cart has to follow within the cell. The first color in the flag row corresponds with the color of the first workstation in the routing. The second color corresponds with the color of the second workstation, and so on. In summary, two types of colored flags are used to guide the material flow: the colored numbered flag indicates the product type, while the colors on the (unnumbered) flags specify the routing (Figure 13.2).

Figure 13.2Details of the flag system on a material cart: A red numbered flag, and three flags (green, yellow, grey) indicating the routing.

When this initial step of coding the parts with the flags is completed, the carts are placed in the middle of the stove cell. The cell is now ready to process the parts. After a cart is processed at a workstation, the corresponding colored flag of the workstation is removed by the operator and the cart is put back in the middle of the cell. The cart is now available for processing at its next workstation. With this system, operators can easily spot which parts they have to process without consulting a computer system or printed shop orders. They just have to look at the colored flags to know which carts are available to be processed next at any given workstation.

The production control system in the stove cell is not a POLCA system. Nevertheless, it shares several important characteristics with POLCA that help to explain its success:

• As in POLCA, the work in process (WIP) in the cell is strictly limited by the finite and small number of available carts. Limiting the WIP has several advantages for the cell. Material piling up can lead to all kinds of waste (searching for material, larger walking distances, damage, and so on), and so all these wastes are avoided. Also, limiting the WIP in turn limits the lead time of jobs in the cell.

• It is a highly visual system that is easy to understand and use for the operators.

• The order information (the status indicated by the flags) is always up to date. So, the production control system doesn’t rely on outdated information, which used to be the situation.

As a result of the simple rules, the workflow is completely self-steering. Employees are in control of the process, which means they feel more involved. Detailed planning of the separate workstations by a planner, and supervisory tasks like shifting workers between the work stations, have become unnecessary. Interestingly, the quality of the products has also improved significantly: scrap and complaints have been reduced by 60%.

The switch to a single production cell meant that material movement on the shop floor was also reduced and stoves could be made in batches of 15. The lead time per batch decreased by a massive 85%, from around four weeks to just three days. The packing process in the cell was also observed to be twice as fast because the products were directly on the right pallet and weren’t stored in the warehouse waiting to be picked.

The success of the stove cell and its lead time and near-perfect delivery performance resulted in more products being ordered by the customer, and Provan now makes eight models of stoves. Even so, the on-time delivery performance is still 100% even with the many more models, and the promised delivery time to the customer has been shortened from five weeks to three weeks. And, of course, one of the best results for Provan’s management has been that instead of tripling the warehouse space (or more), the stocks of parts have actually been eliminated, freeing up 600 square meters of space for Provan to expand its production operations.

Expanding to the Rest of the Factory with POLCA

After this initial success, Provan looked for further opportunities to implement QRM cells. However, this proved much harder. After analyzing its product mix and routings, Provan concluded that there were limited opportunities to form additional cells focused on clear product families. In addition to the stoves cell, an assembly cell for medical treatment couches was put in place, and a third cell was dedicated to prototyping work. These three cells represented about 50% of the work volume of the company. For the remaining 50% of the work, since Provan served as a subcontractor for a wide variety of customers and products, its environment represented a typical job shop. The volume and work content of jobs at Provan was rapidly varying, which made it difficult to dedicate work to particular cells. So, after the success of the first three cells, the company was left with the complex challenge of managing the workflow for the rest of its jobs.

For these remaining high-variety jobs, Provan suffered from several problems on its shop floor: information at the work centers was often already outdated even by the time the production orders were distributed to the shop floor; planning and re-planning activities formed a heavy burden for the company; supervisors had to run around the shop floor to expedite orders; and priorities were shifting continuously. Finally, despite all the follow-up and rescheduling, the necessary parts to start an order were often missing. As a result, the order information was not trusted by the operators.

To solve these issues, Provan undertook several steps. First, the planning was simplified by organizing the shop floor into manufacturing cells in which similar operations were grouped together. In the past, planning and order routing was done at the detailed level of individual machines. Now, the planning is done at a higher level of cells, and operators in the cells are given more responsibility so that they are able to handle the work themselves in a more flexible way. (For more insights into this approach, also see Appendix C.) Second, the high variety of jobs can be accommodated by routing jobs through different cells based on their needs.

Third, Provan decided to implement a POLCA system to control the work flow between the cells. As the team at Provan started looking into POLCA, it had several concerns about implementing the system with physical cards, and, after some deliberation, the team chose to implement an electronic version of POLCA. Several reasons motivated the option of an electronic solution:

• After analyzing the typical variety of orders going through the shop, it was calculated that a large number of different routings were possible between the cells. This would result in a huge number of possible POLCA loops and cards and it was felt it would be difficult to maintain and update all of these.

• The work content of the subcontracted jobs had large variations: some orders might take only a few minutes, while others represented several hours or even days of work. This meant that to smooth out the work flow to enable POLCA to work effectively, Provan would need to implement a quantum rule (see Chapter 5). However, with the large variation in jobs, as well as new and custom jobs that frequently arrived, someone would have to analyze the quantum for each job and the operators would need to deal with varying numbers of POLCA cards attached to each order. This would complicate the implementation for both the planners and the shop floor operators.

• The team felt that with the large number of cards and loops, physically returning the cards to the previous cell would be seen by the operators as a waste of their time, which would undermine their acceptance of the system.

After considering various electronic options, Provan chose to implement a system called Digital POLCA from PROPOS software. The system includes one screen for each cell on the shop floor (Figure 13.3), which displays the information on jobs at or destined for that cell, and a central screen for the supervisor and planner with more detailed information (see Chapter 12 for the background on PROPOS as well as more details of the system). Implementing an electronic version also offered the advantage that the system could be customized for Provan’s needs, as described next.

Figure 13.3PROPOS screen at one of the cells.

Customizing the System for Better Use of Scarce Resources

In addition to the POLCA functionality, one straightforward customization of the PROPOS system at Provan was to use the system for recording the time spent on each job at each cell. However, a more elaborate customization proved beneficial for Provan’s unique situation.

Lead times at Provan were already quite short because of the flexibility of its resources and management’s belief in planning for spare capacity. This meant that with the operation of the POLCA control rules, there would typically be very low WIP levels in the factory. At times, some of the cells were standing idle waiting for orders. Of course, this can happen naturally when there are simply no orders left for these cells, and if you plan for spare capacity this is bound to happen occasionally. However, at other times management noticed that orders that would use the downstream idling cells had been released to the shop floor, but these orders were delayed at upstream cells that happened to be overloaded at this point of time because of the unplanned variability of the incoming work. Again, in some cases this might be acceptable, and is just part of the dynamics of high variety operations, but this situation was particularly concerning to Provan’s management when the idling cells were staffed by welders. Because welders are hard to get in the labor market in Belgium, it is not easy to have extra capacity in welding. At the same time, Provan’s expertise in welding gave it an edge in getting orders. So, any idle time at the welding stations was seen by management as a lost opportunity.

These factors motivated Provan to ask PROPOS if they could come up with a solution for this situation. The aim was to have a solution that did not involve complex shop floor scheduling algorithms because of the drawbacks of such systems (see Chapter 3), and also kept to the main framework of the POLCA rules.

As is the case in a typical POLCA implementation for high-variety jobs, upstream cells service multiple downstream stations. At certain points in time, some of the downstream cells were idle or almost idle, while others still had sufficient orders to keep them busy for a while. This situation offered the possibility to prioritize orders at upstream stations that were destined to go to stations that were soon to experience a shortage of work. In fact, POLCA offered an easy way to find the stations with a low work load! A low work load at a downstream station corresponds to a large number of available POLCA cards at the upstream station. So, the algorithm just had to prioritize orders based on how many POLCA cards of the downstream stations were available. This was still done under the constraint of respecting the authorization dates. In other words, if multiple jobs at an upstream station were authorized, then the system would pick the job based on this prioritization. This added an elementary layer of logic that was simple enough for everyone on the shop floor to understand and buy into, while maintaining the simplicity and clarity of the POLCA system.

The enhancements provided by PROPOS have been received well at Provan. The Digital POLCA system now improves the flow of work in both these situations: when there are a lot of orders at Provan, the POLCA logic limits the work at every station, prevents excessive WIP, and also ensures that upstream cells work on jobs that will be needed by downstream stations, using the normal POLCA logic. However, when there is lower demand, the added logic helps to ensure that critical cells such as welding do not unnecessarily run out of work.

Results: The People Side

The first benefit that Provan noticed after the POLCA implementation was the increased transparency on the shop floor. In the past, only the planner and the supervisors were (more or less) aware of what was going on. If a problem arose, they had to be consulted, which unavoidably led to additional delays. With the transparency of the system, everybody had access to the same, real-time, and thus up-to-date information. Over time, everyone found that this information was typically accurate, so then the operators started to trust the system, and this provided a huge benefit of buy-in from the shop floor.

Now, if an order is in danger of being late, a warning appears automatically on the upstream production screens, and since the shop floor personnel now trust this information, they can start to think how they can resolve this situation. The system further supports this by displaying the reasons for delay (such as waiting for parts from suppliers).

The system also had a major impact on the activities of the supervisors. The supervisors had serious doubts before the implementation: they objected that the system might not be as smart as an experienced supervisor. They also feared that their jobs would be even harder with two added duties: trying to follow the system, and at the same time trying to correct its deficiencies. However, with the system now in place for a while, these same supervisors actively endorse the system and are glad to have its support! They have experienced that the system has liberated them from many of their cumbersome tasks. Instead, these supervisors can now focus their energy on coaching and developing their people, cross-training, and other improvement activities, and leave the logistics issues to the PROPOS system. This is a better use of the experienced resources represented by the supervisors, and also reinforces Provan’s competitive advantage through ongoing personnel development and other continuous improvements.

Results: The Commercial Side

Provan got an early signal during this journey that it was on the right track when a major customer re-sourced some of their products from a low-cost country (LCC) to Provan. This is particularly impressive when you realize that Belgium has among the highest labor costs in the world after you factor in all the benefits and taxes in addition to the wages. In fact, the customer calculated that because of Provan’s extremely short lead times and high delivery reliability, they could agree to a price from Provan that was 11% higher than the price from the LCC: Provan’s lead times and delivery record provided the customer with substantial savings in indirect costs such as inventory, warehousing space and personnel, rush freight charges, rescheduling, and so on. This was a classic win-win situation with lower total cost of ownership for the customer while at the same time supporting a reasonable profit margin for Provan.

Thanks in large part to its successful implementation of QRM and POLCA, in 2015 Provan received the prestigious “Factory of the Future” award. This distinction is awarded by the two Belgian industry federations Agoria and Sirris to companies that are best prepared for the future. This award typically places these companies among the global leaders in their sector. Some of the most striking results that were mentioned by the award jury included the significant reduction in lead times, and a tripling of the “added value” of the company (a metric used in Belgium that measures the difference between the annual revenue and the purchase costs) over a five-year time span.

About the Author

Pascal Pollet is a principal engineer at Sirris, the collective center of the Belgian technology industry, where he started doing research on POLCA in 2005. Pascal has been pioneering the introduction of Quick Response Manufacturing (QRM) in Belgian industry and has helped many companies to radically shorten their lead times with QRM. Pascal obtained a Masters degree in Electrical Engineering from Ghent University in 1995, and is a member of the European QRM Institute.

Ben Proesmans is the co-founder and owner of the Belgian company Provan (www.provan.be), a fast-growing metal-working supplier. Ben is one of the QRM pioneers in Belgian industry, and is also affiliated with the European QRM Institute as an Ambassador. His company was recently awarded the prestigious Factory of the Future Award as a recognition for its pioneering work.