Business-to-Business EC Activity in Small Manufacturers

Jonathan A. Morell, Ph.D. and Mark E. Brown, Ph.D.

Published by The Journal of Electronic Commerce Volume 12, Number 2, 1999, reprinted with permission

This project began as an effort to test the consequences of electronic commerce (EC) in a defense manufacturing supply chain. In particular the emphasis was on the business value of EC for the small companies within such chains. Six suppliers to the Oshkosh Truck Corporation (OTC) agreed to participate. Data were collected on the full range of their EC activity, regardless of whether or not OTC was the trading partner. Over the two year period of the study, (1996 - 1997) we collected three types of data: monthly quantitative metrics on a variety of business functions, as-needed quantitative data to assess unforeseen events, and structured interviews with technical and business managers.

Our primary finding was that even though the suppliers engaged in a limited range of the EC that was potentially available to them, the EC they did use yielded important benefits. Table 1 provides a pictorial overview of a typical range of EC for a small manufacturer, and identifies where the actual activity took place during the two years we studied these firms. (The darker the cell coloring, the more data we were able to collect.)


Product Data Exchange in Electronic Form
The most dramatic impact of EC involved the exchange of CAD files in digital form as e-mail attachments. (See Figure 1 for a pictorial overview of these effects.)

This simple innovation had major advantages in terms hours available for value-added engineering, efficiency of communication, and engineering lead time. In addition to these direct effects, we were able to trace the impact of reduced engineering lead time to the size of inventory, and from inventory to overall profits. These benefits derived because of the specific business climate in which the ECOTS suppliers operated.

• The companies work mostly as job shops. While they certainly have a lot of repeat business for stock parts, they also received a great many orders for new parts or for modifications to existing parts. In these cases some engineering is required.
• Engineering is on the critical path to order fulfillment because manufacturing cannot begin until the engineering is completed.
• Customers commonly accompany orders with three 2-dimensional drawings, one for each view (front, side, top). Manufacturing requires integrating these three drawings into a single 2?dimensional (isometric) drawing.
• Processing the drawings is time consuming.
• Ambiguities in specifications are common with paper drawings because their quality is degraded by activities such as scaling down when copying, faxing in poor resolution, or folding to fit in envelopes.

To reduce time and error two of the ECOTS suppliers began exchanging CAD files as email attachments with those of their customers who were able and willing to do so. We could work out the detailed consequences of this exchange because the companies were generous enough to provide us with information on:

• Number of orders per month with customers which required engineering.
• Number of drawings per order.
• Time needed to convert from multiple 2-D to a single 2-D, for drawings in both electronic and paper form.
• Error rates in interpreting specifications that arrive on paper.
• Engineering labor costs.

A word of caution. When considering the advantages we attribute to CAD exchange via email, it is important to realize that while those advantages can be considerable, they are also seldom realized. The calculations in this section are based on savings from the small number of companies with whom the suppliers are using email for CAD exchange, scaled up to the total number of customers with whom they exchange engineering drawings. In fact very few of the suppliers' trading partners had email. While specific numbers are lacking, several of the suppliers reported asking their trading partners about email, and finding that only a miniscule percentage had that technology.

We were able to ascertain how electronic product data exchange affected engineering in three ways: integrating three 2-D drawings into a single perspective, confusion due to ambiguous specifications, and engineering rework. Results of the analysis appear in figure 2.

 While the savings are indeed considerable, it is important to see these calculations as measures of potential and not necessarily of reality. In addition to the uncertainties in our projections from the sample to the population, we do not know how the companies used the freed-up engineering resources. We have no information on what proportion of that time was used to add value.

Lead time

On average drawings in digital form saved the suppliers one half day of lead time. This is a small savings in relation to the three to ten week lead times that are typical for the companies we worked with. Even this small lead time reduction, however, had consequences that could be traced to savings in inventory carrying costs, and from there to the bottom line (figures 3 and 4).

While the percentage of profit increase was less than one percent, it is noteworthy that the small capital investment needed for email systems can affect engineering processes in ways that have discernable effects on profits.

Email for Efficient Communication

The most profound impact of email emanated from its ability to provide very fast transmission of graphics that contained none of the ambiguity that is often present in paper or faxed drawings. This ability provided companies with a range of possibilities:

• decrease overall order fulfillment time -- by getting information to suppliers more quickly
• decrease express delivery costs -- by avoiding expediting costs when fast communication is required, and
• provide more accurate information to suppliers -- by delaying transmission of orders for a few days and using the time to check on requirements.

Also with respect to email, it is important to bear in mind that besides CAD file transfer, email can make other contributions to improving the time and cost of communication with customers. The two major uses of email were order tracking and expediting. These firms received frequent queries from customers about order status. Before email the common procedure for the recipient of such requests was to wander from department to department until the order was located. With email the time and effort of searching was replaced by a single one-to-many email message, and while waiting for a reply, the customer service representative was free to do other work. In addition the customer with the request got much better service. Respondents estimated that they could reduce their labor by about 2 hours per incident, and response time to customers by about 1.5 hours. We could not obtain quantitative estimates of how frequently these searches took place, but respondents agreed that such searches were common occurrences.

The second use of email for communication with customers was to facilitate the negotiations needed to resolve expediting requests. The ECOTS suppliers received many requests to move up delivery dates, and it was often unclear if the request could be honored, or if a different, mutually agreeable, date could be determined. By eliminating phone tag, email made these negotiations much more efficient. Respondents could not provide quantitative estimates of the value of email for this purpose, but they did agree that email technology was of great help.

Electronic Data Interchange

While none of the suppliers used integrated EDI, five got to the point of testing transactions with OTC, and three used EDI for routine business activity with OTC. In addition two of the firms used (or were about to use) EDI with customers other than OTC. In each of these cases being EDI-capable played an important role in the company's ability to get or keep the business. (See Table 2 for a summary of the suppliers' EDI activity).

A combination of data from the firms (transaction volume) and contextual understanding (EDI and labor costs) allowed us to learn a great deal about the conditions under which EDI is cost beneficial. (Analyses are based on data for existing EDI activity, projected to suppliers' total business volume.) Figures 5 and 6 provide different perspectives on those conditions.

Figure 5 looks at one year payback in terms of per-transaction cost for a single trading partner. Of the five suppliers for whom we were able to collect data, only one has a transaction volume that could justify EDI. The outlook is more sanguine when the cost justification horizon extends beyond one year (Figure 6). Then, three of the five firms could justify EDI within three years. Because of the rate of technological obsolescence, 3 years approaches the limit for a cost justification time horizon. Thus given their transaction volume, some companies cannot justify the investment.

Further insight comes from the fact that in Figures 5 and 6 the suppliers are plotted in terms of their transaction volume only with OTC, i.e. the plots assume that a company has only one EDI trading partner. Because transaction volumes increase rapidly as trading partners are added, a multiple trading partner business environment can greatly expand the range of settings where EDI can be cost-justified.

The finding that EDI makes sense for only a limited number of ECOTS-like firms is confirmed by an entirely separate analysis done by OTC for its own purposes. While that analysis used different assumptions about costs, and focused on a different sample of suppliers, their conclusion was similar to ours, that the break-even point is about 33 transactions per month, and that: "24 of 192 vendors for the period 9/1/97 through 11/30/97 had more than 33 PO line items per month".

Before leaving the subject of EDI justification, its important to remember that considerations other than transaction costs play a part in implementation decisions. Chiefly, those reasons are process improvement and business necessity in the form of a customer demand for EDI. If process improvement is a priority, integrated EDI can play a critical role in developing the powerful information systems that are needed for efficient, fast-response, agile manufacturing. If business necessity is the motivator for EDI, the critical issues are whether the customer is important enough and insistent enough. If he is, EDI is easily justified as one of the costs of doing business that are required to maintain customer satisfaction. Under these circumstances the only question is what kind of EDI to implement -- the cheapest that will satisfy the customer, or something that can be built on should the future reveal a need for higher transaction volume, greater transaction complexity, or integration. If increased demands on EDI are likely, initial implementation should be with technology that is compatible with the likely increased need.

Impediments to adoption of EDI

The findings summarized in figures 5 and 6 are based on the use of integrated EDI, i.e. that form of the technology that can decrease time and labor cost for data entry. These savings were not realized because while several of the firms talked about a desire to do integrated EDI, none put serious effort into making it happen. We could not get systematic data on why integration failed, but anecdotal evidence suggests several reasons. First, OTC began transmitting EDI quite late in the project. As a result the suppliers accumulated less experience with the technology than might be necessary to make the jump to integration. Also, fewer transactions were implemented than had been planned. Thus was lost whatever justification may have developed based on accumulated transaction costs. In addition, two technical issues came into play. First, the structure of data sent by OTC (multiple line items within a single purchase order) did not match the "release" oriented system of one of the suppliers. Resolving this mismatch would have been a considerable task. In another case a supplier had been waiting over a year for a promised new release of MRP system, and was reluctant to integrate EDI until the new system was in place.

World Wide Web

Data on Web use were limited, but sufficient to demonstrate how the Web can be of value to ECOTS-like firms. Table 3 summarizes findings for the five situations where respondents could clearly identify business value for their use of the Web.

That value falls on a continuum of "impact specificity", i.e. the precision with which respondents could identify value. At the least specific end of the continuum is "assessing competition", i.e. a business planning function in which specific Web-based information is only one of many inputs needed to make a decision. In the middle of the continuum is marketing activity. Here respondents could identify specific actions taken on the basis of Web-based information, but were unable to identify bottom line consequences of their actions. Finally we have "finding suppliers" and "checking product specifications" - activities that are wholly within the work domain of an individual Web user, and whose value those users could quantify based solely on their personal experience. All along the continuum, whether quantifiable or not, there was general agreement that the small effort needed to learn to use the Web was well worth the effort.

Summary

For two years we observed (and assisted) five small manufacturers adopt EC in ways that fit their business needs and technological capabilities. Their choices were simple and opportunistic. Simple in that no major changes in business process or system integration were required. Opportunistic in that change came from individuals acting within their personal discretion, without need for deliberate planning or close coordination among several different departments or functions. The result of this activity was limited change that had very great potential.

• There could be a significant increase in engineering resources and decrease in lead time if a greater number of trading partners had the ability to exchange product data in digital form.
• The efficiencies of email for problem solving and order tracking was limited because only a few of their trading partners had email capability.
• EDI capability was attractive to customers, but lack of transaction volume and problems with internal systems made integration not cost-effective
• Web use was limited to applications that did not require integration of Web sites to internal systems, or a customer base that could actually do business via the Web.

Throughout, it is clear that two factors stand between small companies and the realization of EC's potential, inability to scale-up to larger numbers of transactions and trading partners, and inability to integrate internal functions.

It is also clear from our findings that small firms can employ EC in creative and productive ways. If they see value, and have the ability to act, they can change quickly. Many small firms are, in fact, agile. Action by a single trading partner, however, limits the use of EC, which is, after all, a process of interaction among companies. For its full realization, EC must be implemented in an environment of cooperation. Most small firms lack power relative to their trading partners, and thus cannot force those partners to accept EC business practices. Similarly, "resistance from trading partners" is a well known impediment to EC, even when powerful companies are driving the effort. To make the system work, trading partners must fashion models of EC which benefit both, and employ implementation strategies that make adoption feasible.