Impact of one piece flow on economic data quality and shop floor data collection.

ABSTRACT

This article focuses on the basics of one-piece flow production as a market-oriented just-in-time (JIT) production system that strives to fulfill the requirements of total quality management (TQM), reduce product changeover time, increase productivity, reduce work-in-process inventory, and eliminate all sources of waste. The article outlines the impact of implementing one-piece flow manufacturing method on an organization's manufacturing agility. Aside from any productivity gains realized and financial gains associated from carrying the financial burden of excessive inventory, organizations can benefit from the improved quality (credibility) of the economic data it generates. These improvements are realized due to the reduction in total inventory primarily through dramatic improvements in WIP (work in process inventory) levels. The process of data collection from the shop floor represents an additional benefit of implementing a one-piece flow production system. A reduction in data entry requirements for production reporting and inventory tag system maintenance occurs once one-piece flow is implemented. Improvements in the quality of data contained in monthly, quarterly and annual financial reports and reductions in labor hours to maintain production and inventory control are very significant in today's competitive and dynamic manufacturing environment. This article will outline specific examples of how one-piece flow contributes to improving the quality of economic data and improving the shop floor data collection process.

1. INTRODUCTION

The capability of a manufacturing system to be flexible enough to respond to the changes in the marketplace and customer's needs is a crucial element of survival. One essential prerequisite to such a capability is the ability to reduce the machine changeover time to zero, to be on-line with the suppliers and users, and to have goods brought to the site on a daily basis and shipped immediately. Another requirement for competing in a tight marketplace is to eliminate all sources of waste. These changes are symbolized in a relatively new type of manufacturing system, known as one-piece flow production system, in which the machinery are smaller-size general-purpose machines that are lined up in a U-shape cell and are managed by only a few operators. The one-piece flow (OPF) production system is a hybrid of the best of all the traditional production systems that insists on waste reduction as the base for improvement (Ohno, 1988).

The literature is filled with articles on the advantages of a zero-defect production system (Houshyar and White, 1996). They all endorse the same concept: build a thorough inspection system into the production line, so that parts are conveyed, processed, and inspected one at a time, and make sure that the human element of the production/inspection system is trained to appreciate the significance of maintaining high standards of quality in his/her work-cell.

One-piece flow is accomplished by implementing a flow of material through production operation one piece at a time without being placed in containers for transport to subsequent operations (Sekine, 1992). The parts are optimally produced in "U" shaped cells by one operator. The operator is responsible for checking his/her own work and assuring that the required quality level is met prior to sending the part off for further processing, assembly or packaging and shipping. The parts are transported to upstream operations via carts, conveyors or simply passed on a table leading into an adjacent cell. Figure 1 illustrates a one-piece-flow layout in which there are four cells, 12 production operations and a maximum of 16 parts in the system. In a one-piece flow system product is pulled through the production operations as ordered by the customer. Parts are not produced to maximize equipment utilization and as such they do not remain in the staging areas waiting further processing or waiting to be sold.

[FIGURE 1 OMITTED]

By implementing one-piece-flow organizations realize dramatic reductions in work-in-process inventory (Robinson, 1991). This reduction in inventory is realized due to:

1) Parts are not being stored in containers (unit loads) at operations while they are being processed. Instead one piece at a time is processed in cells and ideally only one piece is in transit between operations.

2) Parts are made as they are ordered. Batches or lots of parts are not staged between operations waiting to be scheduled and then to be processed.

Figure 2 depicts a typical production operation making four different parts, three production lines, with each line performing the same operation. As the parts are processed, they are placed in containers (in this case 50 per container are used as the unit load). Comparing the parts in inventory in Figure 1, it is seen that there are 16 parts vs. the 1200 in Figure 2 (a reduction in WIP to 16/1200 = 1.3%). This equates to a 98.7% reduction in the amount of inventory contained in WlP. Any fluctuations in the value of WIP now become insignificant compared to the other more stable inventory components namely finished goods and raw materials.

[FIGURE 2 OMITTED]

Reductions of total inventory, which is composed of raw material, WIP, and finished goods of ten times or more are not uncommon when one-piece flow is implemented. In what follows, the impact of this reduction in inventory on the quality of the economic data that an organization generates will be examined.

DISCUSSION

In what follows the impact of the following main factors with an impact on the uncertainty of inventory value will be briefly discussed.

1) Scrap

2) Raw material value adjustments

3) Inventory adjustments: both point of operation value and quantity discrepancies

4) Obsolescence, shrinkage and perishable costs.

Scrap: WIP value is an estimate that is mostly affected by the amount of scrap contained in it waiting to be discovered as it is processed. WIP contains parts, components and subassemblies that will be processed directly into finished goods, reworked, repaired and sent to finished goods or they will be scrapped out. If the scrap rate is high for a given financial reporting period, then the reduction in value associated with the WIP, directly reduces the bottom line financial performance of the company, in addition with one-piece flow scrap rates are lower than typical batch production methods as less inventory is between inspection points. This improves feedback and correction cycle of process control. The timeline for detection and correction is shorter and thus fewer parts are at risk of becoming scrap. Thus if the initial scrap rate is at a lower threshold and fewer parts are at risk due to the lower inventory level, the total compound effect to uncertainty in the value of inventory results in a much lower dollar value of uncertainty associated with the existing WIP.

Figures 3-8 illustrate the relationship between uncertainty in inventory as a function of inventory component value: Raw materials, WIP or finished goods. The distribution of this uncertainty is also outlined. As the amount of WlP is reduced the total contribution that scrap makes in the level of uncertainty will decrease. Once One-piece flow is implemented and optimized uncertainty in inventory value is minimized as WIP is reduced together with raw material and finished goods inventory.

FIGURE 3. DISTRIBUTION OF INVENTORY AND ASSOCIATED UNCERTAINTY IN A
TRADITIONAL BATCH SYSTEM.

Inventory value distribution and levels of uncertainty in a
traditional batch system

                         Inventory Component

                  Raw   WIP   Finished   Total

Inventory value   100   100     100       300
Uncertainty        1     10      1         12

Raw material valuation: Some products produced in manufacturing facilities have the value of the raw material used in the products adjusted on a periodic basis to compensate for market value of the raw material. If there is a lot of material value tied up in WlP inventory while the adjustment takes place, then the effect can have a sizeable impact on financial performance. By minimizing WIP (and thus total inventory value) by implementing OPF, this source of variation of the total inventory value will be minimized.

FIGURE 4. PROBABLE DISTRIBUTION OF UNCERTAINTY IN A TRADITIONAL BATCH
PRODUCTION SYSTEM.

Inventory
adjustments           20%

Obsolesence,
Perishable or
Inventory shrinkage   15%

Raw Material Value
adjustments           15%

Scrap                 50%

Note: Table made from bar graph.

Inventory Adjustments: Inventory accuracy is always an issue of considerable importance in a manufacturing facility. The value to place on a product at various points in the manufacturing process is always a controversy as actual costs vary, yet the assigned value is typically held constant over time. By using one-piece flow and U-shaped cells to process the part through several manufacturing stages, fewer discrete value assignments need to be made. In addition with fewer parts in inventory, any incorrect assumptions or outright errors associated with the value of a part at any point in the process is minimized.

FIGURE 5. DISTRIBUTION OF INVENTORY AND ASSOCIATED UNCERTAINTY IN A
ONE-PIECE-FLOW SYSTEM

Inventory value components and levels of uncertainty in a one-piece
flows system

                         Inventory Component

                      Raw   WIP   Finished   Total

Inventory Component   100    10     100       210
Uncertainty            1      1      1         3

FIGURE 6: PROBABLE UNCERTAINTY DISTRIBUTION IN A ONE-PIECE PRODUCTION
SYSTEM.

Inventory
adjustments     10%

Obsolesence,
Perishable or
Inventory
shrinkage       15%

Raw Material
Value
adjustments     65%

Scrap           10%

Note: Table made from bar graph.

Inventory discrepancies found while trying to reconcile physical inventories are very time consuming (as multiple recounts are typically required to verify the discrepancy) and they represent significant variation in inventory values. If a discrepancy occurs in an organization that has implemented one-piece flow, its impact on financial data will be much smaller than the impact that an inventory discrepancy might have had on the organization had it used batch production system.

FIGURE 7. DISTRIBUTION OF INVENTORY AND ASSOCIATED UNCERTAINTY IN A
OPTIMIZED ONE-PIECE FLOW SYSTEM

Inventory components and level of uncertainty in a optimized
one-piece flows system

                         Inventory Component

                      Raw   WIP   Finished   Total

Inventory Component    50    5       50       105
Uncertainty           0.5   0.5     0.5       1.5

FIGURE 8. PROBABLE DISTRIBUTION OF UNCERTAINTY IN AN OPTIMIZED ONE
PIECE-FLOW SYSTEM.

Inventory
adjustments           10%

Obsolesence,
Perishable or
Inventory shrinkage   10%

Raw Material Value
adjustments           75%

Scrap                  5%

Note: Table made from bar graph.

Obsolete or Perishable Inventor and Inventory shrinkage: Inventory adjustments caused from parts becoming obsolete, perishing and shrinkage (theft, lost or damaged) are minimized when one-piece flow is implemented. This is mainly due to the reduction in WIP produced for the product ordered, and parts are made to order and not placed in inventory until sold. Additionally the risk for parts perishing or becoming obsolete and reducing the value of inventory in a one-piece flow environment is much lower when compared to a traditional batch production environment.

Shop Floor Data Collection: Most manufacturing facilities use tags to identify what type and number of parts are in the container, what processing has been completed, and what processing remains to be done. Routing tags are always a nuisance. They need to be printed, filled out, attached to the container, collected, stored, and eventually disposed of. Tagging errors are always a concern during quality audits, as every manufacturing facility has experienced missing tags and tagging errors. Compounding the tagging nuisance is the fact that plant personnel need to be trained, and continually retrained on proper use of the tags. Production and inventory control personnel are constantly looking for tags, correcting tagging errors and redesigning or creating tags for new or improved products.

One-piece flow can be implemented without tags; the parts are usually fully completed within self-contained cells. Tags become unnecessary except for finished goods, raw materials and components, and non-conforming material/product. This dramatic reduction in tagging represents a direct cost savings in terms of printed material (the tags), labor (filling out, creating, collecting and auditing the tags), and disposal costs (land-fill).

Production reporting in a one-piece flow environment becomes a much simpler task when compared to a batch production environment. In reference to Figure 1 where 12 machines are divided into four cells each containing three operations, it is seen that the parts undergo all three operations and the three machines essentially operate as one. Moreover, only one production report per cell would be required. Contrast this to Figure 2, where those machines operate independently in three lines for which 12 production reports would be required. This reduction will lessen the impact that production-reporting errors can have. With less reports being generated, operators will have the opportunity to be more accurate as they can focus on one report for the cell vs. trying to maintain two or three reports for the three machines that they could be operating in a batch production system. Since the data entry load is reduced, the number of data entry errors should be minimized as more attention can be placed on fewer reports. Additionally discrepancies should be easier to reconcile with the fewer production reports involved. The labor savings observed through fewer transactions occurring links one-piece flow manufacturing methods with "Lean Accounting" principles as well (Henderson and Larco, 2002 and Maskell, 2002).

3. SUMMARY

Aside from the immediate benefit that lower inventory has on cash flow we have outlined additional benefits that can be realized by organizations implementing one-piece flow. With fewer tagging errors and production reporting errors using one-piece flow, the quality of the production data collected from the shop floor will translate into more credible financial data due to the lower levels of uncertainty regarding the data. With lower levels of uncertainty in inventory value resulting in lower WIP and total inventory levels, financial data will be of higher inherent quality. Without the noise that a lower quality financial data can introduce to the measurement of financial performance, improvements or declining performance can be identified as it is occurring. Finally, the financial anomalies discussed regarding inventory value become less a factor in a one-piece flow production system.

4. REFERENCES

(1.) Bruce Henderson and Jorge Larco, Lean Transformation: How to change your business into a lean enterprise, Oaklea Press, Richmond VA, First Edition, 2002.

(2.) Azim Houshyar and Bob White, Effectiveness of Japanese Transplants: One-Piece Production, 19th International Conference on Computer and Industrial Engineering, Miami, Florida, March 3-6, 1996.

(3.) Taiichi Ohno, Toyota Production System: Beyond Large-Scale Production, Productivity Press, Cambridge, 1988.

(4.) Alan Robinson, Continuous Improvement in Automation, Productivity Press, 1991.

(5.) Kenichi Sekine, One Piece Flow, Productivity Press, Cambridge, 1992.

(6.) Brian H. Maskell, President, BMA Inc. and Bruce Baggaley, Senior Partner. Lean Management Accounting, http://www.maskell.com/LAMPath4.htm, 2002.

Azim Houshyar, Western Michigan University, Kalamazoo, Michigan

Gordon Peters, Western Michigan University, Kalamazoo, Michigan