Just-in-time has become one of the most-discussed management topics of the last decade. This chapter discusses the elements that make it effective.
Introduction
Taichi Ohno, a former shop manager and eventual vice president of Toyota Motor Company, is the individual most credited with the development of just-in-time. Just-in-time (JIT) is a term coined to describe the Toyota production system, widely recognized today as one of the most efficient manufacturing operations in the world. Today JIT is widely used in the automotive industry everywhere. But it is not limited to the automotive industry; many firms use JIT or some of its elements. It has created something of a revolution in how operations are managed.
Basic Elements of JIT
Just-in-time is more than an inventory control system. It is a philosophy and integrated management system based on the concept of eliminating all waste. Just-in-time production is also known as lean production. The intention of just-in-time production is to produce only what is needed, when it is needed.
Waste has a very comprehensive meaning in just-in-time systems. Some examples of waste are
Many techniques are used for eliminating waste in a just-in-time production system. Ideas come from employees working on continuous programs of improvement. There are also other common elements of just-in-time that define the philosophy and management system:
Some of these terms may be familiar to you from coverage in other chapters. Other terms are new concepts. Each of the elements will be detailed in the remaining sections of this chapter to illustrate their fit in the just-in-time philosophy.
Flexible Resources
Flexibility is the key to eliminating waste, like excess or obsolete inventory and worker idle time. The following resources are the source of such flexibility:
Cellular Layouts
Cellular layouts should be familiar to you from previous chapters. Recall that jumbled flow patterns are a problem in process (functional) layouts where similar machines are grouped together. Cellular layouts eliminate the jumbled flow pattern by:
Worker routes and volume decreases:
The Pull System
The pull system of production is an important method of minimizing work-in-process inventory. In general, the pull system is a simplified method of self-regulating a system of production, while a push system is a method of maximizing individual production rates.
Recall the line balancing problems we reviewed in Chapter 7. When several workstations are combined into a production system, some stations have slightly more work to fill the time available, and some have slightly less. For example, an assembly line may have 4 work stations, with the following average times to complete work: 30 sec, 25 sec, 40 sec, and 30 sec. In a push system, each worker is encouraged to complete as much work as possible. Consider what will happen at each workstation in the assembly line we just described. The first worker will "push" his work to the next station at a slower pace than the second station can use it. The second workstation will have 5 seconds of idle time between each unit. The third station, on the other hand, will not be able to keep with all the production arriving at the station. A unit will arrive every 30 seconds, but it takes 40 seconds to complete work at that station. What happens as each worker works as hard as they can? Work will build up at station 3. Before long, a huge amount of work-in-process is built up because each worker is working at an individual best, not at a system pace.
In a pull system, the system pace is determined by the slowest workstation in the system. A worker cannot pass on any work to the next station until the next station has passed its work on to its subsequent station. In our assembly line example, the worker at station 2 would have to wait about 10 seconds after completing a unit before passing it onto the next station. You may be thinking: what a wasted resource! In fact, it would be impossible to produce any faster than the slowest workstation in the system anyway. The only thing that happens when workers produce at their own individual pace is that work-in-process builds up. You should see that a pull system is much better than a push system because it keeps the process more visible and the area neat. Work-in-process inventory hides the process problems and creates messes that cause confusion. In a pull system, workers can see when a line is highly unbalanced and make changes to correct the problem. For example, workers might balance out our assembly line by shifting some work activity back to station 2, away from station 3.
Kanban Production Control System
In the example we just described, workers would be able to use a pull system with a minimum of difficulty because they sit next to each other and see when the next station is empty and ready for more work. When workers are in different areas or cannot see each other, some signaling system is needed to indicate when workers are ready for more work. The signaling system used is called kanban.
Kanban is the Japanese word for card. It is the "visible record" used in a pull system. It works as follows: a bin arrives at a workstation with work from a previous workstation. The worker removes the work from the bin and sends the kanban back to the previous workstation when the work in the bin has been reduced to a certain reorder point. When the kanban arrives back at the previous workstation, it is a signal that more work can be passed on. Work is passed on with the same kanban, and it arrives just as the next station completes the work in the previous bin. This cycle repeats over and over again. This method is a single kanban system. A small amount of inventory is kept in the system to allow for the transportation time between stations.
A dual kanban system and different types of kanbans are illustrated in Figure 15.6.
It is easy to get caught up in the technical aspects of kanbans and lose sight of the objective of the pull system, which is to reduce inventory levels. Kanbans simply provide the means for signaling when work needs to flow. The kanban system should always be kept as simple as possible. A kanban system should always encourage the continual reduction of inventory. We can see how that occurs by examining the formula for determining the number of kanbans need to control the production of a particular item.
Determining the number of kanbans:
where
Example:
A bottling workcenter processes an average of 150 bottles per hour. If one kanban is attached to every container, a container holds 25 bottles, it takes 30 minutes to receive new bottles from the previous workstation, and the factory uses a safety stock factor of 10 percent, how many kanban containers are needed between the bottling process and its predecessor process?
L = 30 minutes = 0.5 hours
dL = (150)(0.50) = 75
C = 25 bottles
S = 10% dL = (.10)(75) = 7.5
N = 
= 
=
3.3 kanbans containers
Start with a worker that has three containers (# 1, #2, and #3) of bottles, and the predecessor container-filling station that has 1 container (#4) ready to go when signaled by a kanban. The worker removes bottles from container #1 at t = 0 and sends the container back for refill. The roundtrip refill begins instantaneously at t = 0, and container #4 is sent on its way when container #1 returns (let's assume at t = 15 minutes; both parts of the trip taking an equal amount of time). During the roundtrip time for container #1 (30 minutes), the worker completes the other containers (#2, #3, and #4) at 10 minutes intervals and sends each back for re-fill. The containers now arrive at 10 minute intervals. The first container arrives back at the bottling station just as container #4 is emptied. It has idle time of 10 minutes before it is emptied a second time. A container can be removed to reduce inventory by the container amount, but then production and movement rates must be perfect every time.
In the example we just reviewed, container #1 leaves the bottling process for a refill at t = 0 minutes. It returns with the refill at t = 30 minutes because it takes 30 minutes for the roundtrip. Try to track each container mentally using t = 10 minute intervals. Answer the following questions to make sure you understand the kanban flow. Remember, at t = 0, the bottling process begins on container #1. At t = 10, container #2 is emptied at the bottling process. Write down the times that each container is emptied at the bottling process if it helps you see what is going on. Then write down the times each container spends, in-transit, on the re-fill trip.
Producing in small lots has many benefits, including
Large lots result in batches of work-in-process inventory. Conventional wisdom in just-in-time thinking is that inventory hides problems. The thought of reducing inventory to very low levels can be worrisome, but JIT philosophy posits that it is better to expose problems so that they can be fixed. Figure 15.7 illustrates this concept.
A goal of JIT is to reduce lead time, which is made up of four components:
Quick Setups
Setups are adjustments that must be made on equipment or processes each time an item is changed from one model to another or one product to another. Setup time can be very lengthy -- often hours long. When setups are long, manufacturers often want to produce a large number of the same item before changing to another. The concept of long setups does not work well with small lot production.
Shigeo Shingo is well-known for his SMED (single-minute-exchange of dies) principles, which were developed to reduce setup times. For example, Shingo reduced the setup time on a 1,000 ton press from six hours to three minutes using the following principles:
Guidelines for reducing setup time include:
Figure 15.8 illustrates some common techniques for reducing setup times:
Uniform Production Levels
Uniform production levels help moderate the amount of inventory in the system and avoid the use of excess overtime. Kanban systems can handle fluctuations of 10%, but any more than that creates pressure on the system to create excess inventory. Production is leveled by the use of better forecasting techniques and the use of mixed model sequencing.
Mixed model sequencing was illustrated in Chapter 6, but it will be reviewed here once again.
Example:
If Toyota receives a monthly demand estimate of 1200 small cars, 2,400 midsize cars, and 2,400 luxury cars, how should the models be produced in order to smooth production as much as possible?
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Solution:
First, convert monthly demand to a daily schedule by dividing by the number of days in a month. The result is a daily production schedule of 40, 80, and 80 of each model, respectively, per day.
Models should be sequenced by finding the ratio of each model volume to the smallest model volume. The ratios of midsize and luxury cars to small cars are both 2 to 1. Two midsize and luxury cars should be produced for each small car.
A sequence of L-M-S-M-L repeated 40 times per day maintains the proper mix of models.
Quality at the Source
Quality must be extremely high in a JIT system because there is little inventory to buffer against quality mistakes. A JIT system should have a zero defect policy that seeks to identify quality problems at their source. Workers, not inspectors should be responsible for quality. Worker responsibility for quality requires the following components:
Total Productive Maintenance
Two basic types of maintenance are
Total productive maintenance (TPM) seeks a higher degree of maintenance than preventive maintenance. Total productive maintenance combines the practice of preventive maintenance with the concepts of total quality -- employee involvement, decisions based on data, zero defects, and a strategic focus. TPM requires management to
Supplier Networks
Just-in-time purchasing and supply has developed rapidly. Trends in supplier policies include:
Benefits of JIT
The benefits of JIT are similar to those of advanced manufacturing technology, but they are achieved through reduction of waste and productive management of human resources. In essence, JIT achieves the four strategic objectives of manufacturing simultaneously -- low cost, high quality, high flexibility, and quick delivery. These overall benefits come from
JIT Implementation
Just-in-time production began in Japan in the 1970’s and spread to the U. S. in the 1980’s. We can make these general observations about JIT:
(Discussion points: Different cultural values, different regard for space, group orientation of Japanese, economic shift to higher productivity in U. S., more flexibility in U. S. culture, market emphasis in U. S.}
{Discussion points: Both require innovation; technology is more costly and risk, but eliminates human discontent; JIT can treat workers like machines, but can also optimize human potential; workers get satisfaction of contributing ideas}
JIT in Services
JIT was developed and implemented in manufacturing industries, but the applications for services are increasing. Many services base competition on speed and quality, and the philosophy of JIT meshes well these objectives. Some service applications of JIT include: