Week 3

A process flow (or flow, for short) is a sequence of processes and stockpoints through which entities pass in sequence

is the rate of good (non-defective) entities processed per unit time. Tons of steel produced per day, cars assembled per shift, or customers served per hour are examples of throughput measures. Note that it is important not to count defective product, which will need to be reworked or remade, as part of throughput

is the time between the release of an entity into a routing and its completion. In a flow that produces subcomponents, cycle time will measure the time from when raw materials are drawn from a stock to when the component is placed in an intermediate crib inventory. In a flow that produces final products, it will measure the time from when the entity starts down the flow to when it is placed in finished goods inventory or shipped to the customer.

measures the inventory in a flow. Generally, this does not include raw materials or finished goods inventory. However, for flows that cut across multiple processes, it may include intermediate crib inventories. While there is some flexibility in defining the start and end of a flow, it is important that the same definitions be used for both WIP and CT in order to make these consistent.

• You can compare flows to converyors

• The basic behavior of a conveyor is described by two parameters:

  • The rate at which items are placed on the front and removed from the end of the conveyor

  • The time time it takes an item to go down the conveyor.

Every workstation has something to work on. When the entitity is finished, it can go to the next workstation without waiting. This special WIP level, which results in both maximum throughput and minimum cycle time, is called the critical WIP

• Extremely larges batches or extremely high variability

• Batching is due to setups and material handling issues, while variability is the result of many factors, including quality, reliability, staffing, scheduling, and others. In practice, what this means is that either large batching or variability problems can push the performance of a process flow toward that of the worst case

• Balanced flow

  • All stations in the flow have the same capacity.

• Single server stations

  • All stations consist of one server, and hence can only work on one entity at a time. If a station had multiple servers working in parallel, then when one server experience a delay, entities can continue to flow through the other servers.

• Moderately high variability

  • Process times of entities at every station are so variable that the standard deviation of the process times equals the mean process time. CV of all processes is 1

Comparison of performance against theoretical capability

Is a standard of comparison based on performance of another system

• Collect four parameters:

  • bottleneck rate (BNR)

  • Raw process time (RPT)

  • Average work-in-process (WIP) level

  • actual throughput (TH)

• If TH > THPWC, then the flow is in the good region; otherwise it is in the bad region.

if all stations in a line process and move entities in batches, then the waiting time caused by batching will be the sum of the waiting times at the individual stations. The batches propagate from one station to the next and so does the waiting they cause.

• Consider a station that experiences both flow variability (i.e., variability in the interarrival times of entities to the station) and process variability (i.e., variability in the process times at the station).

• This will generate variable interoutput times.

• Because these will be the interarrival times to the next station, variability in them will cause queueing delay at the downstream station.

• If the station is highly variable, the outputs will also be highly variable. If the station has low process variability, the outputs will also be of low variability. If input has high variability, the process low, output will still be high if utilization is high

• If the variability of a non-bottleneck is improved, the entire chain improves, because the flow variability to a bottleneck is reduced and improved.

• Conclusion of picture: When station utilization is high → queues start → station variability = output variability. When utilization is low: Input variability = output variability

• Improve system parameters

  • Increase the bottleneck rate BNR or decrease the raw process time RPT

  • Speeding up the bottleneck increases BNR, while speeding up any nonbottleneck process reduces RPT. Processes can be sped up by either adding capacity (e.g., replacing a machine with a newer, faster one) or via more subtle means such as improving reliability, yield, staffing, or quality.

• Improve performance given parameters

  • The two primary means for moving away from the worst case and toward the best case are to (1) reduce batching delays at or between processes by means of setup reduction, better scheduling, and/or more efficient material handling; and (2) reduce delays caused by variability via changes in products, processes, operators, and management that enable smoother flows through and between stations

• Inventory

• Can also be buffered by time (if there is no inventory)

• Maintain backup machines = capacity

• They can be used simultaneously

• Variability in a production or supply chain system will be buffered by some combination of inventory, capacity, and time.

• A mix is needed

• Design of the physical production environment is an important aspect of management policy

  • Certain machines are needed to realize a strategy: Burger King wanted more options, thus rapid cooking was needed

• Different operations systems can be used for different products.

  • McDonalds → only products with high turnover will be stocked

• The appropriate operations system for a given application will change over time

  • Because both the physical environment and business strategy fluctuate and/or evolve over time, the operations system needs to adjust as well.

A particularly important aspect of buffers is the extent to which they are flexible.

• Flexibility reduces the amount of buffering required in a production or supply chain system.

• All buffers are costly, so minimizing them is key to efficient operation of production and supply chain systems.

• Flexible inventory

  • Stock that can be used to satisfy more than one type of demand.

• Flexible capacity

  • Capacity that can be shifted from one process to another. A common example of this is an operator who has been cross-trained to perform multiple tasks so that he/she can float to stations where work is piling up

  • a flexible manufacturing system, which can switch quickly from producing one product to another.

• Flexible time

  • A production system that quotes fixed lead times to customers (e.g., all deliveries are promised within 10 weeks of ordering) is making use of a fixed time buffer.

How well a buffer compensates for the effects of variability is strongly influenced by its location. The reason is that the throughput, cycle time, and WIP in a process flow are largely determined by the bottleneck process. Therefore, a buffer that impacts a bottleneck generally has a larger effect on performance than one that impacts a non-bottleneck.

• For a flow with a fixed arrival rate, identical nonbottleneck processes, and equal-sized WIP buffers in front of all processes,

  • The maximum decrease in WIP and cycle time from a unit increase in nonbottleneck capacity will come from adding capacity to the process directly before or after the bottleneck.

  • The maximum decrease in WIP and cycle time from a unit increase in WIP buffer space will come from adding buffer space to the process directly before or after the bottleneck

  • the general behavior that WIP buffering is most effective when used at or near the bottleneck is universal.

  • A station with more capacity requires less downstream WIP buffering, while a station with more variability requires more downstream WIP buffering to protect the bottleneck

• Production of goods or services is lean if it is accomplished with minimal buffering costs

  • By thinking of waste as the result of buffers against variability, we can apply all of the principles of this book toward identifying levers to make a production system lean

• What are the Lean implementation steps at Toyota

THE PROCESS LENS

THE FLOW LENS

THE NETWORK LENS

THE ORGANIZATION LENS

Because the “pursuit of waste reduction” definition of Lean lacks any reference to underlying causes of waste, it basically leaves the user to focus on waste that is directly visible from observation of individual processes.

It’s the pursuit of reducing the following types of waste:

1. Defects

2. Overproduction

3. Transportation

4. Waiting

5. Inventory

6. Motion

7. Processing

nothing in the Process Lens or the list of waste categories it has engendered guides us to this conclusion

it lacks guidance on how to diagnose and remedy waste, the Process Lens is best suited to identifying waste that is created directly in the process itself, rather than as a by-product of issues or activities outside the process.

For organizations just beginning their Lean journey, this very simple view of waste reduction can be helpful in attacking obvious waste. But this lens is less helpful in identifying waste that propagates from other parts of the process or beyond it.

The second definition of Lean zeroes in on variability-induced waste. Because the effects of variability are manifested in the flows of people, materials, dollars or other entities, we term this second lens the Flow Lens

Variability is always buffered. Each type of waste has its own type of buffer

A small capacity buffer will lead to larger fluctuations of net-inventory that create either larger stocks or more backorders. The only way to reduce the total amount of buffering (as measured by the product of the buffers) is to reduce total variability

In a production line or a simple supply chain, the number of choices may be sufficiently limited to allow the Flow Lens to achieve much of the potential of Lean.

But most production and service systems consist of many interconnected flows. All of these flows can be represented conceptually as networks of flows, so we label this the Network Lens

The core concept for understanding the behavior of networks of flows, and thereby identifying points of maximum leverage, is that of a bottleneck

The intuition behind this is that the closer a resource is to full utilization, the smaller the fluctuation in demand or capacity needed to overload the resource.

queueing theory also tells us the effect of utilization is nonlinear (exponential), while the effect of variability (as measured by squared coefficient of variation) is linear. This implies that we should first look for ways to reduce utilization (e.g., by adding capacity and/or eliminating unnecessary demand, such as that from rework) and then look for ways to reduce variability (e.g., by smoothing flow into bottlenecks and/or reducing variability in bottleneck processes).

One reason for the gap between the conceptual perspective of the Network Lens and the Lean tools used to address complexity in systems is that much of the academic literature related to bottleneck and queueing analysis assumes stationarity. That is, over the long run resource utilizations are stable, which implies that a unique bottleneck (or at least a small set of bottlenecks and near bottlenecks) will govern system behavior. But in practice, it is possible, even likely, that neither demand nor capacity are stationary

Although a physics focus might be sufficient for a production or service system run entirely by machines (e.g., a true “lights out” factory), the reality is that all business systems involve people. To account for this, we require the fourth and most expansive perspective on Lean which we term the Organization Lens.

A possible implication from the research into psychological biases is that people need more training in probabilistic thinking to deal with problems involving uncertainty. Another conclusion could be that planning under uncertainty in many situations should not be done by intuitive feel at all, but instead should make use of a data-based decision support system

People need to be engaged in the implementation in a way that engages their System 2 thinking and overcomes their System 1 biases.

We are all prone to cognitive biases that can blind us to effective alternatives. Therefore, how we involve people in the search for ways to implement Lean, what data are provided, how questions are posed, and how well people are prepared to think about problems that involve uncertainty, all matter in the effectiveness of bottom-up problem solving.

A low goal makes it easier to avoid a painful loss, albeit at the expense of a reduced likelihood of the satisfaction of a high-level outcome. Moreover, once a goal is set, people have more incentive to reach it than to exceed it. Kahneman (2011) explains that in many cases, people will reduce their efforts once they have reached a specific goal because they see no point going above and beyond.

Think big before thinking small. A system review that simplifies the production network and a flow focus that identifies bottlenecks helps to focus waste elimination efforts on processes that matter to overall performance.

Adjust buffers to facilitate exploration and exploitation. Optimizing variability buffers is a vital part of Lean implementation. But buffers can also be adjusted to reveal the sources of variability.

Pursue physics and psychology in parallel.

Iterate, iterate, iterate. The pursuit of efficiency is a never-ending journey.