Quality in a Virtual Organization
Quality in a Virtual Organization
Background
Charles Darwin sat at his disk on an early Tuesday morning in the first quarter of 1998. The business landscape of the electronic equipment company, GO, for which he worked was drastically different than just one year ago. It was 8 months ago that GO announced the decision to aggressively pursue outsourcing business functions wherever it made sense. The cost advantages and opportunity to concentrate on core competencies was the reason given for the shift in strategy. After all, business at dedicated electronic manufacturers and contract manufacturers (CMs) had been growing phenomenally and many companies currently adopting an outsourcing strategy were singing the praises of what it had done to their business’ bottom line.
In the three months following the announcement that outlined the company’s plan, there was much uncertainty amongst the rank and file. What did it mean for their positions? How would their positions change? What are the core competencies that upper management thinks the business should emphasize? How were they going to interact with outside companies? How would they protect the technology, critical to the success of the business, that is shared everyday with the contractors? Over the course of the following months some of the answers were sorted out and the business was able to get into a groove with the new scenario.
The quality function within the company to which Charles belonged, had many questions as a result of the outsourcing decision. The company traditionally took pride in the quality of its products and used its high quality levels to compete against the competition. The quality function was instrumental in achieving this effect. Previous to the outsourcing decision, the quality function consisted of 3 main areas: quality control, quality data gathering/reporting and external quality. The quality control group held responsibility for the control of quality on the production floor. In addition, the quality control group was responsible for ensuring the manufactured products met safety and governmental regulations in the countries where they did business. The function also had a role in customer satisfaction. External quality individual contributors handled customer satisfaction by owning resolution of any quality/customer satisfaction issues that arose at a customer site.
As manufacturing was outsourced, there would be a significant impact on the quality control group. The members of the quality function speculated on how the decision would affect their group. The functions of managing quality on the shop floor would be transferred along with the elements of production. The responsibility of regulations and safety testing would stay with the GO company but seemed to be an ideal candidate for outsourcing. It was obvious from much of the management communication that the external customer quality function would remain, but who would make sure that internally, the company was upholding the vital quality standards? The answer to this question was Charles Darwin and the internal quality individual contributor role. The internal quality individual contributors participated in new product development activities and other business operation activities to ensure that quality was being considered in daily decisions. (See Exhibit 1 for the current quality organization structure). Perhaps the most uncomfortable question of all was how the new quality organization would impact the product quality if various aspects of the product were outsourced.
Charles opened up his appointment calendar. Tuesdays were his busiest day. First a program meeting for a new project, Aegis, that had just completed the first run of prototypes. The topic of the meeting was a go/no-go decision for a second prototype run. Then at 10am he had a meeting regarding the transfer of final assembly from the company lines to a new CM. At the end of the day, was a monthly quality roundup evaluating the quality of the business’ products from manufacturing to full assembly to field performance.
9:00am Aegis Program
Prior to the Aegis Program meeting, Charles ran into the lead senior engineer, Rick, in the hall. Although this specific issue was at the top of the agenda, Charles could not help but ask Rick if the second prototype run would be able to start on time. Since the second run could not go forward until the first prototypes yield satisfactory operation, Rick confessed that he had received bad news last night from Excetech, the company to whom the main control chip design had been outsourced. Two months ago, GO’s management determined that the control chip function was straightforward enough for GO to feel comfortable with outsourcing its design to Excetech. Excetech was chosen because they had other high performance control chips on the market and offered a reasonable price for design development. Since the decision, Excetech had designed the chip and selected a foreign semiconductor foundry to produce it. After production, the chips were sent back to GO’s rapid prototype CM for incorporation into the Aegis product producing the first 6 prototypes. After one week and lots of overtime, four of the 6 prototypes were found to be non-operational. The lead engineers were convinced that the main control chip was to blame so the 4 faulty chips were sent back to Excetech for testing. Last night, Rick had received word that Excetech’s testing showed all 4 chips to be operational according to the evaluation tests they performed at their site.
The team assembled for the Aegis program meeting consisted of representation from nearly all the functional groups within the business, from R&D to marketing to support and manufacturing. At the meeting, Rick relayed the situation to the rest of the team. Afterwards, he began to theorize on what the possible problem might be saying, “I have a feeling that these problems are related to poor quality.” Rick went on to explain that Excetech subcontracted its design to a foreign foundry for manufacture. Going to a foreign foundry in itself is not a problem, however, the foundry Excetech selected had a questionable reputation for providing quality output. Additionally, Rick had heard rumors of priority conflicts between Excetech production and the foundry’s other customer business. “The only problem with this supposition, is that Charles and I discussed this possibility with the folks at Excetech, and they told us that the possibility of poor manufacturing was minimal. They reassured us that all 4 chips had been designed per specification, were built correctly and now operate correctly on their reference board.”
The program manager reasoned, “Well if they were all built and tested to specification and the chips passed even after they were removed from the prototypes… perhaps the issue is some mismatch in specifications.” Rick defended that the needs for the chip were relayed to Excetech in a document generated by himself and a materials engineer. After the document was delivered, a conference call was held to answer any of Excetech’s questions. Rick felt confident that Excetech was set to deliver a product consistent with GO’s expectations.
As the discussion on this issue was wrapping up, Charles reminded everyone that it was crucial to GO’s reputation to attain high quality for this product. If quality was not built into this product, sales would suffer and sales of other products may suffer as well. “I agree,” answered the program manager, “but it is also vital for us to stay on schedule. If we don’t make it to market on time, we know that our competitors will be out with their version of this product. If we come out late, sales may never pick up like we expect.”
“So what can we do to get to the bottom of this issue?” The R&D manager began to relate that a lengthy investigation into this quality issue made it all but impossible to get to market before the competitors’ products. Furthermore, he stressed that his group was in a difficult situation. The Excetech engineers’ dedication to teamwork has been fading ever since his team decided that the chip malfunction was due to a problem with Excetech’s design. The R&D manager felt that his group could figure out the problem if they had the design database for the control chip, but under the contract terms and conditions, Excetech could not share the database with GO.
The program manager started to recount the options the team faced. If they decided to go ahead with a second prototype run, 30 prototypes would have to be built at a cost of $400,000 due to the low yield. The program would need a minimum of 10 functional protos after all possible debugs to keep the program on track. Another option would be to hold off and wait for a resolution of the Excetech problem, but any delay would prevent the program from making it to market on time.
The team probed the possibility of using another vendor that could supply a chip with very similar functionality. The R&D manager said he knew of another potential chip on the market. He thought that the chip could possibly serve as a replacement but circuitry external to the chip would have to be added by engineers on his team. The R&D manager pointed out that there was a relatively good chance that the chip would work, although performance would suffer due to the external circuitry requirements and it was unclear exactly what the external circuitry design would entail.
The choices facing the team were not easy. All at the meeting, Charles included, were torn between the different options. Charles understood that the product had to get to market so the business could be successful, however he needed to see good performance from the Excetech chips in order to allay his quality concerns. To close the meeting, the program manager decided to take these options before the executive board for their consideration and decision.
10:00AM Subcontracting Final Assembly
The next meeting at 10 would have a new set of issues that Charles would have to participate in resolving. Manufacturing was making a critical decision on subcontracting out final assembly of the product. On one hand this made sense to Charles, as corporate was pushing to consolidate production at international manufacturing hubs, and the impressive tax savings and lower labor costs would reduce the product cost by 20%. However, Charles was very worried about the quality implications. He expected that hidden costs and quality deviations could easily eat up the tax and labor savings. It would be a tough meeting, as Jordi Bartolome, the Manufacturing Manager, would need Charles’ buy-in to launch the subcontracting.
The Manufacturing department had explored what outsourcing possibilities were present in the market and decided that there were three alternatives. The “in-house” option was to manufacture in the Singapore division of the company. Being part of the company meant that communication would be easier and that processes would be either in place or easy to leverage. However, the Singapore division did not offer substantial savings on labor or overhead. Their selling argument was primarily the tax savings, so the cost reduction of going to the Singapore division would only be 10%.
The other alternatives were with CMs. One CM, Omni Industries, had been identified as making an especially good fit with the company since its founders were originally from the Singapore division. Given their trim infrastructure, Omni could compete on both the labor and the tax savings. The Singapore division itself and other company divisions were already working with Omni. However, Charles’ division had never worked with them in the past.
Two other international CMs in Singapore, Flextronics and SCI, were also candidates. Both CMs already worked for the division manufacturing circuit boards and electronics assemblies. They, like Omni, had lean overhead structures, and offered the full 20% reduction in product cost, when including the tax advantage. The division’s long-standing relationship with both companies was very good. There were few quality issues, and the ones that came up were identified internally on the final assembly line.
Charles’ own opinion was that the product line should not be subcontracted. As the company quality pundit he could see that the move was fraught with risks. The Aegis program was already evidencing the quality problems in subcontracting. He was also considering the external supplier quality issues he had been studying, ranging from Ford’s problems with Firestone to Boeing’s problems with subcontracted hydraulics systems. The company’s goals were to reduce production costs, but they also had a Cost of Quality reduction goal of 15% per year. His experience told him that quality would not improve in the subcontractor, but to the contrary, it would worsen. He made Cost of Quality calculations and found that a 15% increase in failure rate would offset the product cost reductions that Manufacturing expected. What if the quality issues were worse? What if they were faced with a product recall?
At the meeting, Charles’ and Jordi’s views collided. Jordi was convinced that going to Flextronics in Singapore was the only way that the company would be able to achieve the cost reduction goal for the year. He submitted the past success of subcontracting electronics subassemblies to them. Charles presented his concerns and the Cost of Quality calculations he had made. He also noted that the externally made subassemblies always passed through the company for final assembly, where they would be tested as part of the product. If they outsourced final assembly, the customer would be the first to test the subcontractor’s quality!
Jordi and Charles agreed to work out a quality assurance plan for the subcontracting scenarios. Jordi made it clear that Manufacturing wanted to move forward on subcontracting final assembly, and that together they would need to take the necessary steps to guarantee quality. The final decision would be made next week at the functional staff meeting. Charles left the meeting as troubled as before.
What steps could they take to guarantee quality?
What option was the best for the company?
3pm Quality Review
The quality data gathering/reporting group within the quality function had historically gathered production data from the shop floor. In addition to production data, the group also acquired field failure data and reported external failure rate. The group also computed the Cost of Quality, the important metric of dollars spent on customer quality as a percentage of revenue. As subassemblies were outsourced, the group no longer had responsibility for the shopfloor, however they still held the responsibility of reporting production data.
The quality data group, after outsourcing began, developed a plan to gather data from the various CMs it would do business with and then make that data available internally throughout the company. After a month, the quality group was able to gather data from 3 of the 5 manufacturing centers and 2 of the 3 assembly/ship centers. The groups that abstained were concerned that the data would be ‘used’ against them. For instance, they felt that if they were late on a delivery, the quality charts would be looked at and if quality levels were worse than historical levels, GO management might falsely blame problems on poor supplier quality.
On a monthly basis, data from the various CMs was delivered to the quality department. The quality group would then break out the data by product, a Pareto chart of the failures would be created, and the failure rates would be plotted against goals. The failure rates were compared against goals set during the proto phase, based on failure rates of historical products of similar technology. Once all the graphs and reports were prepared, they would be distributed amongst the business management for review.
At the quality meeting, the standing agenda was to review the quality book assembled by the process. Attendees of the meeting included the manager of the quality data group, the internal quality individual contributors, the manager of product engineering and the quality functional manager. During the meeting questions regarding apparent issues were directed to the person who had the most direct impact on product quality of anybody in the room, Jordi B. The meetings were a frustrating experience for Jordi because the production data was a month old, external failure rates for products were distorted in early life and the details needed to provide accurate answers on product performance were known engineers working for him, but they weren’t at the meeting.
Charles had received feedback from Jordi regarding some of these frustrations. Charles planned to raise questions at the meeting to develop a better working model for the quality review meeting. Who should get this information? Shouldn’t production data be the responsibility of the contract manufacturer? Is this information useful? To whom and why? At this point it seems that the report and information is circulated and discussed mostly by the management. Finally, how is the business benefiting from all this data gathering? Charles knew that answering all these questions would be difficult, but they needed to be addressed if the meeting would effectively contribute to improving product quality.