Chapter 4

How to Map a Process

Measuring Processes: Lead Time and Cycle Time

Lean focuses on reducing process lead time, and so the first step in Lean is to measure the lead time of the process as it is currently executed. For the process described in Figure 1 (see chapter 1), we need to measure the average amount of time that elapses from the time that raw wood is cut in the first step until the time when an end table is completed. Determining total process lead time is conceptually straightforward: We need to measure the average lead time required for each process step and then add all those times together. For the first step, where raw wood is cut to length and width, the elapsed time through this step is the time from when a particular piece of raw wood is delivered to this cutting step until that piece is moved from the cutting workstation to the next step in the process. The time that a piece of wood stays at a workstation will vary from piece to piece, and we represent that varying lead time using its average value. In particular, the average could be calculated by taking several measurements and averaging them. Tagging multiple pieces of wood in some manner and tracking them through the process so that lead time can be tracked is tedious, and so we will offer an alternate method for measuring lead time in the sidebar titled “Alternate Method for Determining the Lead Time of a Process Step.” For now, however, think of the average process step lead time as the average of many measurements.

Note that raw materials are often delivered in batches in manufacturing processes, which, for the cutting process step, would be the case when more than one piece of wood is delivered at one time. Deliveries of a batch of wood would be expected in a furniture-manufacturing process in which transporting wood by the pallet-load would reduce material handling time in comparison with delivering wood one piece at a time. Processing can occur in batches in service and administrative processes as well, examples of which are given in the sidebar titled “Batch Processing in Service and Administrative Processes.” Computing the lead time through a process step might be confusing when goods are processed in batches. For example, should the lead time of one piece be measured or the lead time of the entire batch? The answer is that we want to track the lead time for each individual piece of wood. To do so, we must take into consideration all time that a piece spends at a workstation, including the processing time, the time that a piece waits for its turn to be processed, and the time it waits after processing until it is transported to the next process step. When the entire batch is completed before being transported to the next step, one piece must wait for its turn to be processed and it must wait for the completion of the remaining pieces in the batch after it has been processed. There might also be some wait time for the entire batch if other batches are processed before it.

The foregoing discussion makes clear that the lead time of a process step is not necessarily the time it takes for the actual processing of one unit of work at a particular process step. Actual processing time might be only 1 minute, but a piece of wood might spend 2 hours waiting at a workstation, including the waiting time before it is processed and the waiting time after it is processed. The terminology that describes the amount of time that one particular piece of work is actively being worked on at a station is called cycle time (1 minute in this case). Lead time is the total amount of time a piece of wood stays at the cutting step (2 hours in this case). Cycle time and process step lead time are not necessarily equal. Processing work in batches causes the two times to be different, as does waiting for batches that arrived previously to be processed, workstation downtime, and workstations that are not continuously staffed. The only circumstance when process step lead time and cycle time are equal is when raw materials are delivered one unit at a time, the processing step can always immediately start processing the item and is never interrupted by a breakdown, and then the item is immediately taken to the next process step. However, this is an unlikely circumstance.

Notice that if we had measured the lead time of the batch rather than a single part in this example, the lead time for the cutting step would have been the same. However, thinking about the lead time in terms of a single part allowed us to clearly identify the various components contributing to that part’s dwell time, and doing so is important because it is important to distinguish between waiting time and actual processing time.

Having measured the lead time of each process step, it might be apparent that we have not comprehended all of the lead time in the furniture-making process in Figure 1. Materials must be transported from one step to the next, which is not reflected in Figure 1. In addition, inventory is sometimes stored between steps in staging areas, warehouses, or somewhere near the next processing station. Our goal is to quantify all the time that work spends in a process, and so we must account for these waiting times as well. Triangles are typically used to denote transportation and storage operations between steps, as shown in the updated diagram in Figure 5. It is also perfectly acceptable to reflect transportation, alternately, as a process step. To measure the lead time of the entire process, we must also measure the lead time of these intervening steps, which we could do in the manner previously described. Now we can add up the lead times for all the process steps to arrive at the lead time for the entire process as shown in Figure 5. The lead time for this entire process is 33,355 minutes.

In studying Figure 5, one might notice that most of the lead time is composed of waiting—that is, periods in which unfinished parts wait in inventory. This observation is made almost without exception. It is a rude awakening when a manager finds that most of the time a product spends in their factory, it is not being actively worked on but rather is waiting to be transported or waiting for the next processing step. Lead time is most often noted in minutes or even in seconds rather than in hours or in days. The rationale for this is, perhaps, that a larger number gives the impression that more improvement is possible and provides the motivation for making that improvement.

Work Units, External Versus Internal Orientation of Projects, and Process Scope

This book will use the terminology work units to describe what is flowing through a process. In the case of Figure 1, the entities that are flowing through the process can be thought of as the tables or, in the beginning of the process, the parts that will eventually be assembled into a table. Thus, in Figure 1, the work units are tables.

Mapping the process in Figure 1 and applying Lean might likely be for the purpose of reducing lead time so that customers can receive their tables as soon as possible, thereby increasing customer satisfaction and, perhaps, positively influencing repeat business. Such a goal could be considered an externally oriented project because the primary benefits of the project are intended to benefit parties external to the company: the customers. Some references on Lean advocate that projects always should be pursued with the external customers’ benefit in mind. However, Lean can also be employed for the benefit of the company itself: Such a project would be internally oriented. To distinguish between externally and internally oriented projects, we will discuss the process depicted in Figure 6, which is a simplification of the process for getting a jet aircraft to the terminal gate and unloaded once it has landed. Improving the process for this process flow might likely focus on reducing the lead time and, in particular, the time it takes to get customers off the plane and on their way once the plane has landed. This goal would be externally focused on the customers, who will either get to their destination sooner or have an increased probability of getting to their next flight on time. This implies that the work units flowing through the process are the customers. The process is typical in that the actions that materially help the customers get closer to getting off the plane take a relatively small amount of time, and the most significant lead time components are constituted by waiting.

Alternate work units might be viewed as flowing through the process steps in Figure 6 as well. For example, the plane also carries luggage, which also needs to get off the plane and to either the baggage carousel or to the next plane. Reducing the lead time of getting the luggage off the plane would also be an externally oriented goal because the primary benefits are likely to accrue to the customer. One might also view the plane as being the work unit flowing through this process. What benefits derive from reducing the time that the aircraft spends getting to the gate and discharging passengers? As Southwest Airlines has demonstrated, reducing the amount of time an aircraft spends on the ground maximizes the time the aircraft is in the air generating revenue.1 Drawing the process map to support an effort to reduce the time aircraft spend on the ground therefore can be viewed as an internally oriented goal: The primary benefits accrue to the airline.

Considering the possibility of using Figure 6 for reducing the lead time of getting luggage processed or getting the plane into the air again raises a question: Would considering the process steps in Figure 6 be sufficient to help us uncover all the possible improvements that would increase the time an aircraft spends generating revenue or reduce the amount of time required to transfer baggage? Focusing on the latter objective, the answer is likely that we would need to consider more process steps in addition to those shown in Figure 6. Other important steps to consider are the unloading of the bags from the aircraft, transportation from the aircraft to the baggage handling system, and the induction of the bags into that system. Of course, we must consider all the waiting time between those steps, which also adds to lead time. If we do not consider these additional steps, then we would likely miss some important opportunities to reduce lead time. If the goal is to reduce the time a plane spends on the ground, then considering the time to get to the gate after landing might be important, as would including other steps like cleaning and restocking the aircraft. Including more steps in the definition of a process is said to increase the scope of the improvement project.

Considering an appropriate process scope is an important part of defining a Lean project. If the process scope is too narrow, then process steps that affect lead time significantly might be ignored. Considering too large a scope makes it difficult to complete a project in a reasonable amount of time and to focus on critical process steps. Thus defining an appropriate scope for a Lean improvement project requires good judgment to negotiate this trade-off.

Waste, and How to Find It

Having mapped the process as in Figure 5, we now must analyze it in order to find the best opportunities for reducing lead time. To do this, we adopt the perspective of the customer who will purchase the product or use the service created by executing the process. It is the customer who decides if our product or service is worth the price we are charging: If a customer finds a company’s offering to be worth the cost, he or she will buy it. We do not want to reduce lead time in any step of the process if it is going to reduce the value of an offering to customers. Note that this is true regardless of whether process improvement is pursued for external or internal benefit: If the value of a product or service to customers is reduced for the sake of better internal business performance, then a company’s existence might be jeopardized. If, however, in executing a process step nothing is done to enhance the attractiveness of the product to customers, then perhaps that step can be shortened or eliminated altogether in order to reduce the overall lead time of the process. To determine which steps provide value from the customers’ perspectives, we ask the following questions for each process step and intervening wait time:

1. Would the customer pay more because this step is performed?
2. Would the customer’s satisfaction be increased because this step is performed?
3. Would the customer choose the product or service generated by this process over a competitor’s product or service because this step is performed?
4. Does performing this step increase the probability of repeat business?

For a step where the answer to such questions is yes, then we say that step is a value-added (VA) step. When the answer to these questions is no, then we have identified what is called a non-value-added (NVA) step. The lead time involved in executing NVA steps is also called waste. Many companies when implementing Lean also borrow a term from the Toyota Production System to describe NVA activity: muda. The waste thus identified becomes the target of opportunity for reducing lead time.

VA and NVA time is recorded in Lean process maps using a zigzag line below the process steps as shown in Figure 7. NVA time is noted on the peaks of the zigzag line whereas VA time is listed in the troughs. This makes it convenient to add up all the NVA times because those data are all aligned on the same vertical level on the zigzag: The total NVA time is noted at the right end of the zigzag. Similarly, the VA time is summed and noted at the right end of the zigzag below the total NVA time.

Note that although we have already computed total lead time for the process, it can be computed by adding the total NVA time and total VA time from the figures computed on the right-hand side of the map. Another use of these sums is to compute the percentage of lead time where something productive is being done to the product or service from the customer’s perspective. Intuitively, we calculate this quantity by dividing the total VA time by the total lead time:

This is called the value-added ratio (VAR), and together with total process lead time, these are the two most important metrics that describe the performance of the overall process.

The VAR for the process in Figure 7 is approximately 11.3% (3,755/33,355). A statistic in this range usually strikes people as being an astonishingly small number. However, many processes, before Lean is applied, have a VAR on the order of 1% or even less. Some manufacturing processes where, for example, large batches are used and a large variety of products are manufactured might have a VAR of much less than 1%.The author, for example, has observed VARs as low as 0.001%. So it is important to map, analyze, and improve a process, or a majority of processing time is likely to contribute nothing to the desirability of a product of service to the customer. Rather, the extra unproductive time adds cost (multiple handling steps and added inventory), reduces quality (more opportunities for damage), and reduces customer satisfaction (because of delayed gratification). Perhaps the most significant value of computing the VAR is to emphasize how much time in a process is wasted. Upon calculating a VAR of 1% or less, it is difficult for any participant in an improvement effort to argue that the process does not need attention. As an exercise, the VAR can be computed for the process in Figure 6: It is approximately 30.1%.

Notice that the peaks of the zigzag line, where NVA time is noted, are aligned with the waiting steps that are between the processing steps. This is indicative that waiting, a vast majority of the time, is NVA, and customers will not pay a company more money for its product or service or choose its product because it spends more time gathering dust in the process. The rare exception to this rule are things such as wine, where aging is an essential part of the process. Still, if these essential waiting times can be reduced, the performance (lead time and VAR) of the process would be increased.

Disagreement can arise in the midst of determining whether a particular process step or component thereof is VA or NVA. Prolonged conversations of this type serve little purpose and are, in a way of speaking, NVA. One way to resume constructive discussion in a team discussion is to agree on the ground rule that all process steps where such disagreement arises shall be considered NVA. This approach has the advantage that it by default considers more steps to be NVA than would otherwise be the case and therefore targets a greater number of process steps for scrutiny.

In some circumstances, some portions of a process step might be VA whereas other components are not. In that case, these components can be segregated. Specifically, the zigzag line below the process maps can have two entries for any process step: The VA portion of lead time can be entered on the lower level, and the NVA component can be added at the upper level, as shown in Figure 8. Figure 8 shows the circumstance where 30 of the total 75 minutes in the sanding step were NVA, such as changing sandpaper on random orbital sanders. This example also shows that waste is found not only when goods and customers wait between process steps but also within processing steps. Intuition leads some to believe falsely that all activities in a processing step are VA time.

Thus far, we have referred to the representations we have drawn of processes simply as process maps. In Lean, however, process maps using the special notation as shown in Figure 7 are referred to as value stream maps. Figure 7 is a fairly complete rendition of a value stream map (VSM), although more detail could be added. The name reflects that our goal is to identify where value is added from the customers’ perspectives and then to eliminate other contributors to lead time. There are many types of process maps, but the value stream mapping format is fairly universal among those who practice Lean.

Categories of NVA Time

The Toyota Production System identifies categories of activities that are generally NVA, and acknowledgment of these is usually included in Lean training. These categories are often referred to as the seven deadly wastes, and the titles of the categories can vary from one Lean program to another. Indeed, we have included an eighth category of waste (the last one on the list) that is sometimes included. If you find activities in a process that fit these descriptions, then they are likely to be NVA:

1. Waiting. This is time when the work unit is waiting for the next VA process, usually between process steps.
2. Overproduction. When the first of two sequential steps processes work units faster than the subsequent step, even for a relatively short period, this is called overproduction, and an inventory buildup results.
3. Inventory. Inventory can be at the end of a process (finished-goods inventory) or between processing steps (work-in-process inventory). Inventory connotes that goods are waiting, either for subsequent processing or for sale to a customer. Overproduction also leads to inventory accumulation.
4. Defects. Defects in goods and services cause excess cost because defective products must either be discarded or reworked if the defect is caught prior to the receipt by the customer. If the customer receives defective goods or services, then they may become dissatisfied and the company will incur a cost if the customer brings this to the attention of the company, possibly in the form of warranty cost. Alternately, the customer may penalize the company by not purchasing additional goods.
5. Transportation. Transportation between processing steps within a factory or between links of a supply chain is NVA. Certainly, transportation between steps is required, but the degree to which it is excessive constitutes waste. Deeming all transportation time to be waste focuses attention on finding ways to reduce transportation. Within a factory, this is often the impetus for moving workstations closer together.
6. Overprocessing. Overprocessing is when processing imparts characteristics to work units that are not valued by customers. These are types of activities that are likely to be thought of as VA within the organization that produces a good, but researching customer expectations, needs, and desires would reveal that the customer would not pay more for a good or service with these characteristics.
7. Wasted motion. Just because a person is in motion and working hard does not mean that anything of value is being done for the customer. Reducing wasted motion increases efficiency and helps companies reduce the costs of serving customers.
8. Underutilized human skills and time.2 Whenever a worker or manager is idle, this represents time that could be spent on activities that generate value for customers. Reorganizing process flows can sometimes utilize employees to fuller advantage. This category might also be extended to underutilized equipment.

One might notice the close relationship among waiting, overproduction, and inventory: Both waiting and overproduction result in an inventory of goods to build up, or whatever work unit is flowing through the process.

These categories of waste were originally developed with manufacturing processes in mind. However, they can easily be applied to administrative and service processes. For example, transportation can be thought of as the exchange of data between the parties at different steps of a process when services are being provided (e.g., processing applications, processing invoices, or processing insurance claims). Overprocessing can be thought of as providing some aspect of a service that is not valued by a client, such as providing a redundant hard copy of a document when an electronic version would have sufficed. Defects, rather than being mistakes made in producing physical goods, are errors made in the execution of administrative processes and the delivery of services. Overproduction in administrative and service processes results in a backlog of work at one processing step, just as in manufacturing, but the work backlog in administrative processes and services is often in the form of an electronic representation of the work that needs to be done rather than a physical pile of work-in-process (inventory). The translation of waiting, wasted motion, and underutilized resources from manufacturing processes to other processes is fairly apparent.

Drawing Value Stream Maps

Value stream maps can be constructed using a variety of methods, each with its own advantages and disadvantages:

1. By hand. This is easy to do and makes maps easy to change. Flip charts can be utilized for larger scale drawings and for involving team members. Alternately, representing each step on its own piece of paper and pinning them to the wall allows for easy rearrangement of steps when a team member remembers a step that has not yet been included in the map. There is nothing wrong with this approach at all, except that if the value stream map needs to be presented, it will most likely need to be rendered electronically so that it is compatible with a PowerPoint presentation.
2. Excel. Excel is widely available and offers the advantage of accumulating VA time, NVA time, and total lead time using a simple formula. The graphics capability of Excel to draw rectangles and triangles is convenient. Other icons require the user to construct templates.3 Excel generates a reasonably aesthetic map, although other approaches (PowerPoint and Visio) would be considered by many to be more aesthetic.
3. PowerPoint. It is easier to do graphics with PowerPoint than with Excel and the resulting value stream map is already in the format needed for presentation. One disadvantage is that VA time, NVA time, and lead time need to be computed manually, but this is not a large amount of work. It is convenient to have an Excel spreadsheet in the background to store process step lead times and recompute the total lead time when data change.4
4. Visio. Visio creates perhaps the most aesthetic value stream map. In addition, recent versions of Visio come with a template with the shapes needed for value stream maps. Data can be specified for process steps, such as cycle time, changeover time, number of workers, percentage yield, and so forth, which can automatically appear on data graphics that are associated with the rectangles for process steps. Disadvantages include the fact that fewer people are familiar with Visio than with either Excel or PowerPoint, and there is a bit of a learning curve to use Visio to its full advantage in generating value stream maps. In addition, Visio is not included with the Microsoft Office suite and therefore represents an additional investment.
5. Electronic value stream mapping (eVSM). This is an add-in to Visio and it provides additional functionality not included with Visio. For example, it allows data about process steps to be specified and tabulated more easily, and it makes it easier to sum up the total lead time of a process by downloading process data to Excel or entering formulas directly in Visio. It includes other functionality, such as creating spaghetti diagrams. The disadvantage of eVSM is that it represents an additional cost. (See http://www.evsm.com.)

Careful Observation of the Process

Taiichi Ohno is credited with encouraging a method of waste identification called “standing in a circle,” in which managers would literally stand in a circle drawn on the floor of a factory for hours on end.5 This exercise has many potential lessons.

First, concentrating on a process intently for a prolonged period allows many instances of waste to be observed. Once the mission of a process is established in terms of the value it provides to customers, then any action or inaction that does not accrue value to customers is waste. Of course, the found waste can be categorized according to the seven deadly wastes. The list that one might generate from observing a process for 6 hours, and the subsequent categorization, can impress upon an organization that significant improvement is possible. This exercise can, therefore, help to convince an organization not familiar with Lean that it might be a good method to adopt. An example of an activity similar to standing in a circle is described in a case about the Deaconess-Glover Hospital in Needham, Massachusetts.6 In that case, a researcher followed a nurse, observing her activities for 1 hour. A review of the activities revealed that approximately two-thirds of the time the nurse was not contributing to their core mission of helping their patients to become healthier. That is, two-thirds of their activities were non-value-adding. Such a figure should grab a manager’s attention and cause them to ask why. In this case, much time was spent searching for items that did not have a standardized location, asking questions about patient status when information should have been readily available, and wasting effort with needless travel between patient rooms and other locations in the ward.

Standing in one place, if it is done at only one step of a process, is the antithesis of how a Lean project is prototypically approached, which normally follows a process from start to finish in order to draw a value stream map. It is likely that an entire process cannot be viewed from one vantage point. It is more likely that portions of many processes might be observed from one spot. While the value stream map allows waste reduction opportunities to be prioritized in light of the entire process, observing waste from one vantage point might still unearth fruitful opportunities. If done from many vantage points, however, intently observing a process could result in a complete value stream map and a good understanding of the process.

One Lean practitioner learned about another benefit of standing in a circle: It can teach one to question whether every action is value-adding or not.7 Learning to instinctively ask questions like the following activates a healthy skepticism that does not take for granted that any activity is inherently value-added:

1. What is this person doing now?
2. Why is this person doing that?
3. Where is this person going?
4. Why is this person going there?
5. What is this person waiting for?

Workers and managers can fall into the trap of thinking that the activities required by the current process are value-adding, when they are, in fact, not value-adding. Questions that start with “why,” “where,” “how,” and “what” challenge that complacency.

Remember, Lean is a data-driven methodology. Whether the standing in a circle is used or not, somehow a process must be observed or the experience of people involved in the process tapped in order to discover how a process is executed. This knowledge is the basis for improvement and allows for the identification of waste.

What Next?

Having found waste, or NVA time, with a value stream map, we must next eliminate it. Many tools exist within the Lean tool kit for that purpose. We describe some of those tools in part 2. Applying Lean tools to eliminate some of the NVA time allows the process to be remapped as it will be executed after the improvements have been implemented. This rendition of the process map is called the future state value stream map.

Exercises

1. Return to the process that you documented in chapter 1, or select a new process, and do the following:

   a. Draw a value stream map indicating the average lead time for each step or reasonable estimates of each step’s lead time. Also note the lead time for the waiting steps.

   b. Classify all lead time in the process as VA or NVA, and display it on a zigzag line.

   c. Compute the total process lead time and the VAR.

2. Analyze the Harvard Business School Publishing case titled “Deaconess-Glover Hospital (A)”8 by drawing the value stream map for the medication-delivery process beginning at the point in time when a doctor issues a prescription. Make a list of all the waste that you find in that process and potential areas for mistakes to be made.
3. Pick a process or work area in your workplace to observe, or get permission to observe processes in a doctor’s or dentist’s office, a retail store, a fast-food restaurant, or other venue where the process is visually apparent. Watch the operation for 1 hour or more and make a list of wastes that you observe. Then, construct a list of questions you could ask a manager that would determine if there was a purpose to an apparently wasteful practice or would help to uncover the root cause of the waste.