A Just In Time (JIT) system is a considerable departure from the traditional manufacturing system, involving several changes that, in total, are intended to massively reduce the level of waste in a company’s production systems. This also results in significant changes in the types of inventory transactions used. A JIT system has several subcomponents, which are described in this section. A complete JIT system begins with production at supplier facilities, includes deliveries to a company’s production facilities, and continues through the manufacturing plant.
To begin, a company must ensure that it receives products from its suppliers on the exact date and time when they are needed. To do this, the purchasing staff must measure and evaluate every supplier, eliminating those that do not measure up to the exacting delivery standards that will now be used. In addition, deliveries will be sent straight to the production floor for immediate use in manufactured products, so there is no time to inspect incoming parts for defects. Instead, the engineering staff must visit supplier sites and examine their processes, not only to see if they can reliably ship high-quality parts, but also to provide them with engineering assistance to bring them to a higher standard of product quality.
Once suppliers have been certified for their delivery and product quality, a company must install a notification system, which may be as simplistic as a fax machine or as advanced as an electronic data interchange system or linked computer systems, that tells suppliers exactly how much of which parts to send to the company. Drivers then bring small deliveries of product to the company, possibly going to the extreme of dropping them off at the specific machines that will use them first.
So far, we have achieved a process that vastly reduces the amount of raw materials inventory and improves the quality of received parts. Next, we shorten the setup times for company machinery. In most factories, equipment is changed over to new configurations as rarely as possible, because the conversion is both lengthy and expensive. When setups take so long, company management authorizes very long production runs, which spreads the cost of the setup over far more units, thereby reducing the setup cost on a per-unit basis. However, this approach often results in too many products being made at one time, resulting in product obsolescence, inventory carrying costs, and many defective products (because problems may not be discovered until many products have already been completed).
A Just In Time (JIT) system takes a different approach to the setup issue, focusing instead on reducing the length of the equipment setups, thereby eliminating the need to create long production runs to reduce per-unit costs. To do this, a videotape is made of a typical setup, and then a team of industrial engineers and machine users peruse the tape, spotting and gradually eliminating steps that contribute to a lengthy setup. It is not unusual, after several iterations, to achieve setup times of minutes or seconds, when the previous setup times were well into the hours. By taking this step, a company reduces the amount of work-in-process, while also shrinking the number of products that can be produced before defects are identified and fixed, thereby reducing scrap costs.
It is not sufficient to reduce machine setup times, because there are still problems with machines not being coordinated properly, so that there is a smooth and streamlined flow of parts from machine to machine. In most companies, there is such a large difference between the operating speeds of different machines that work-in-process inventory will build up in front of the slowest ones. Not only does this result in an excessive quantity of work-in-process inventory, but defective parts created by an upstream machine may not be discovered until the next downstream machine operator works his way through a pile of work-in-process to find it. By the time that happens, the upstream machine may have created quite a few more defective parts, all of which must now be destroyed or reworked.
There are two ways to resolve both problems: The first is called the “kanban card”, which is a notification card that a downstream machine sends to each machine that feeds it parts, authorizing the production of just enough parts to fulfill the production requirements that are being authorized in turn by the next machine further downstream. This is also known as a “pull system“, because kanbans are initiated at the end of the production process, pulling work authorizations through the production system.
By using this approach, there is no way for work-in-process inventory to build up in the production system, because it can only be created with a kanban authorization. If a kanban must be used to trigger a delivery from a supplier, this can be done with a simple fax transmission, although there is no way of knowing if it has been received by the supplier. A better approach is to add a bar code to the kanban card, which can be scanned into a production terminal, triggering an e-mail order to a supplier; the supplier then sends a confirming e-mail back to the company. The card is then sent to the receiving dock, where it is attached to the supplier’s delivery when it eventually arrives, making the card available for a future kanban transaction when the received quantity is eventually depleted.
The second way to reduce excessive work-in-process inventory and reduce defective parts is to configure machines into work cells. A work cell is a small cluster of machines that can be run by a single machine operator. This person takes each part from machine to machine within the cell, so there is no way for work-in-process to build up between machines. Also, because the operator can immediately see if a part is defective, it is difficult for any but a perfect product to be created by such a machine layout. This configuration has the additional benefit of lower maintenance costs, because the smaller machines used in a machine cell are generally much simpler than the large, automated machinery that they replace. Also, because the machines are so small, it is much easier to reconfigure the production facility when it comes time to produce different products, rather than incurring a large expense to carefully reposition and align equipment. Both kanbans and machine cells should be used together—they are not mutually exclusive. By doing so, a company can achieve extremely low product defect rates, as well as vanishingly small investments in work-in-process inventory.
Before the preceding steps are fully installed, it will become apparent that a major change must also be made in the workforce. The traditional approach is to have one worker maintain one machine, which is so monotonous that workers quickly lapse into apathy and a complete disregard for the quality of their work. Now, with full responsibility for several machines, as well as product quality, workers become much more interested in what they are doing. To enhance this favorable event, the human resources staff must prepare and implement training classes that teach employees how to operate a multitude of different machines, perform limited maintenance on the machines without having to call in the maintenance staff, spot product errors, understand how the entire system flows, and when to halt the production process to fix problems. In short, the workforce must be completely retrained and focused on a wide range of activities. This usually results in a reconfiguration of the compensation system as well, because the focus of attention now shifts away from performance based on high production volumes and in the direction of performance based on high product quality. A major result of having an empowered workforce is that employees are now allowed to stop their machines when they see a problem and either fix it on the spot or immediately call in a repair team. In either case, the result is immediate resolution of the bulk of performance problems.
Finally, the massive changes caused by the switch to a JIT system also require several alterations to the supporting accounting systems. Because of the large number of daily supplier shipments, the accounting staff faces the prospect of wading through an enormous pile of accounts payable paperwork. To make the problem worse, there is no receiving paperwork, because the suppliers deliver parts directly to the production operation, so there is no way to determine if deliveries have been made. To avoid the first problem, the accountants can switch to a single consolidated monthly payment to each supplier. The second problem requires a more advanced solution. To prove that a supplier has delivered the part quantities it claims to have shipped, the accounting system can determine the amount of finished products created during the period and then multiply these quantities by the parts listed on the bill of materials for each product, which results in a total quantity of each part used. The accountants then pay suppliers based on this theoretical production quantity, which should also be adjusted for scrap during the production process (otherwise suppliers unfairly will not be paid for their parts that are scrapped during the company’s production process). This approach also means that there is no need for suppliers to send invoices, because the company is relying solely on its internal production records to complete payments. The types of journal entries required in an advanced JIT system will be posted on my next entry.
The journal assumes no receiving function, with suppliers delivering goods straight to the production floor. This eliminates the need for an initial receiving or quality assurance review transaction, as well as movements into or out of the raw materials warehouse area. Also, because scrap is spotted by the production staff, no separate quality assurance function is needed after production is completed. There is also no transaction to move goods into the work-in-process area, because there is assumed to be too little inventory in this much leaner area to make it worth bothering with the transaction. However, scrap tracking is still necessary, as shown by the first journal entry in the JIT process. The primary JIT transaction occurs immediately after production is completed, where finished quantities are counted and used to create a purchasing liability to suppliers, while overhead is also applied to finished goods, which are shifted to a final storage area. The only other required transaction is for shipment of the goods to customers. There is no need for counting adjustments, because there are essentially no raw materials to count, and finished goods turnover is high enough to leave little inventory on hand. Please note that the process flow and transactions shown in Exhibit 2-5 represent an extremely advanced and streamlined system. In reality, a JIT system may represent a mix of some JIT components and a more traditional system, so additional transactions may be required.
Impact on Waste Costs of Just In Time (JIT) System
A key focus of the JIT system is its relentless focus on eliminating all waste from a system. This can be a waste of assets, in the case of unneeded inventory. It can also be a waste of time, in the case of assets that are unused for long periods of time (e.g., work-in-process inventory held in a production queue). It can also be the waste of materials, such as unnecessary levels of obsolete inventory, defective products, rework, and the like. When fully installed, a JIT system vastly reduces all of these types of waste. When this happens, several aspects of a product’s costs decrease significantly.
For example: by reducing the amount of work-in-process, machine operators can tell immediately if an incoming part from another workstation is defective, and can notify the preceding workstation of the problem before it makes any more parts, which reduces the quantity of rework that must be done. Because a standard quantity of rework labor is often included in a product’s labor routing, a reduction here will shrink the amount of labor cost charged to a product. Similarly, any material that would have been scrapped as a result of improper rework will no longer be lost, so the standard amount of scrap noted on a product’s bill of materials can now be reduced. This also reduces a product’s cost.
The Inventory Journal entry explained:
(1). Write-off scrap/spoilage:
(2) Recognize Inventory Receipt
[Debit]. RM Inventory
[Credit]. Account Payable
Move to finished inventory:
[Debit]. FG Inventory
[Credit]. RM Inventory
Apply overhead Cost:
[Debit]. Overhead Cost
[Credit]. FG Inventory
(3). Inventory Sale
[Credit]. FG Inventory
Overhead costs that are charged to a product will also go down as other types of waste decline. For example: by clustering machines into cells, the materials handling costs that used to be incurred to shift materials among widely scattered machines can now be eliminated. This reduces the amount of materials handling costs that used to be charged to overhead. Also, machine cells tend to reduce the amount of floor space needed, because there is no longer a need for large aisles for the materials handling people to drive their forklifts through; by reducing floor space, one can also reduce facility costs, which will no longer appear in the overhead cost pool. Another form of waste is the quality inspections that used to occur for many machines. Under the JIT system, machine operators conduct their own quality checks, so there is less need for a separate group of inspectors; accordingly, the cost of their pay can be eliminated from overhead. All of these costs (and more) do not directly add value to a product, so they are wasteful costs that are subject to elimination. By eliminating them with a JIT system, fewer costs are left to charge to a product. A key focus of any JIT system is on reducing various kinds of wasted time, so that the entire production process is focused on the time spent actually producing products.
For example: all inspection time is stripped from the system by having operators conduct their own quality checks. Similarly, all move time, which involves shifting inventory and work-in-process through various parts of the plant, can be eliminated by clustering machines together in logical groupings. Third, queue time is eliminated by not allowing inventory to build up in front of machines. Finally, one can eliminate storage time by clearing out excess stocks of inventory and having suppliers deliver parts only as needed. By shrinking the amount of wasted time out of the manufacturing process, a company effectively eliminates activities that do not contribute to the value of a product, which in turn reduces the costs associated with them.
As just noted, the costs of material handling, facilities, and quality inspection will decline as a result of installing a JIT system. In addition, the reduction of all types of inventory will drastically decrease the amount of space required for the warehouse facility. Because all costs associated with the warehouse are assigned to the overhead cost pool, the amount of overhead will be reduced when the costs of staff, equipment, fixed assets, facilities, and rent associated with the warehouse are sharply cut back. Costs will also shift from the overhead cost pool to direct costs when machine cells are introduced. The reason for this change is that a machine cell generally produces only a small range of products, which makes it easy to assign the entire cost of each machine cell to them. This means that the depreciation, maintenance, labor, and utility costs of each cell can be charged straight to a product. This is much preferable to the traditional approach of sending these costs to an overhead cost pool, from where they will be assigned to products in a much less identifiable manner.
Although this change does not represent either a cost increase or reduction, it does increase the reliability of allocation for many more costs than was previously the case.
Despite the shift of many overhead costs to direct costs, there will still be an overhead cost pool left over that must be allocated to products. However, given the large number of changes implemented as part of the JIT system, inventory accountants may find that there are now better allocation bases available than the traditional direct labor allocation. For example, the amount of time that a product spends in each work cell may be a better measure for allocating costs, as may be the amount of space taken up by the work cells that create each product. No matter what allocation system is used, it will be somewhat different from the old system, so there will be a shift in the allocation of costs between different products. In short, overhead costs will decline as some costs are eliminated, while other costs will shift between products as more costs are charged directly to products and the remaining overhead costs are charged out using different allocation methods.
A potentially significant one-time cost that many companies do not consider is the impact of JIT on the cost layers in their inventory costing systems. When a JIT system is installed, there is an immediate focus on eliminating inventory of all types. If a company is using some kind of layering method to track the cost of its inventory, such as last-in, first-out (LIFO) or first-in, first-out (FIFO), then it will find itself burrowing down into costing layers that may have been undisturbed for many years. If so, some unusually high or low costs may be charged off to the cost of goods sold when these inventory items are finally used up. For example: if the current market cost of a piston is $50, but a company has some very old (but serviceable) ones in stock from 20 years ago that cost $20, then only the $20 unit cost will be charged to the cost of goods sold when those units are finally used as a result of clearing out the inventory. Because of the unusually low cost of goods sold, the gross margin will be higher than usual until these early cost layers are eliminated. Because of the lower-of-cost-or-market rule (under which the cost of excessively expensive inventory must be reduced until it is no higher than the current market value), this problem tends to be less of an issue when early cost layers are too high, although the costs charged may still be somewhat different from those for newer layers of inventory. Once all cost layers have been used up, the only costs that management will see being charged to the cost of goods sold are those currently charged by suppliers.
Costing Allocation Differences Between a JIT and Traditional System
The chief difference between the types of cost allocations in a JIT environment and a traditional one is that most overhead costs are converted to direct costs. The primary reason for this change is the machine cell. Because a machine cell is designed to produce either a single product or a single component that goes into a similar product line, all of the costs generated by that machine cell can be charged directly to the only product it produces. When a company completely converts to the use of machine cells in all locations, then the costs related to all of those cells can now be charged directly to products, which leaves costs of any kind left to be allocated through a more traditional overhead cost pool. The result of this change is much more accurate product costs and little debate over where allocated costs should go, because there aren’t enough of them left to be worth the argument.
To be specific about which costs can now be charged directly to a product, they are as follows:
Depreciation. The depreciation cost of each machine in a machine cell can be charged directly to a product. It may be possible to depreciate a machine based on its actual usage, rather than charging off a specific amount per month, because this allocation variation more accurately shifts costs to a product.
Electricity. The power used by the machines in a cell can be separately metered and then charged directly to the products that pass through that cell. Any excess electricity cost charged to the facility as a whole will still have to be charged to an overhead cost pool for allocation.
Materials handling. Most materials handling costs in a JIT system are eliminated, because machine operators move parts around within their machine cells. Only materials handling costs between cells should be charged to an overhead cost pool for allocation.
Operating supplies. Supplies are mostly used within the machine cells, so most items in this expense category can be separately tracked by individual cell and charged to products.
Repairs and maintenance. Nearly all of the maintenance that a company incurs is spent on machinery, and they are all grouped into machine cells. By having the maintenance staff charge their time and materials to these cells, their costs can be charged straight to products. Only maintenance work on the facility will still be charged to an overhead cost pool.
Supervision. If supervision is by machine cell, then the cost of the supervisor can be split among the cells supervised. However, the cost of general facility management, as well as of any support staff, must still be charged to an overhead cost pool.
As noted in several places in the preceding list, a few remainder costs will still be charged to an overhead cost pool for allocation. However, this constitutes a small percentage of the costs, with nearly everything now being allocable to machine cells. Only building occupancy costs, insurance, and taxes are still charged in full to an overhead cost pool. This is a vast improvement over the amount of money that the traditional system allocates to products. A typical overhead allocation pool under the traditional system may easily include 75% of all costs incurred, whereas this figure can be dropped to less than 25% of total costs by switching to a JIT system. With such a higher proportion of direct costs associated with each product, managers will then have much more relevant information about the true cost of each product manufactured.