Taking Risks in Manufacturing
DAVID B. MARSING
One of the biggest pitfalls of a company enjoying a particular market and product dominance is the tendency to avoid risk. This is a particularly serious challenge for those companies, because continuous improvements in manufacturing operations are one of the most effective means of creating and maintaining competitive advantage. But improvements in manufacturing operations are typically associated with change, and with any change there is an element of risk. As companies become risk averse, change, except for historically proven methods, is usually minimized although the pressure for increasing output and reducing costs continues.
Products, processes, and manufacturing plants have life cycles, and, when we are lucky, these coincide with external business cycles. Unfortunately this seldom happens. To achieve higher productivity throughout the entire life cycle of a manufacturing operation requires change. Many of the changes needed to prepare an organization for long-term transitions can take years to work through. Limitations on change can cripple a manufacturing company's ability to improve or even take advantage of future business opportunities. Understanding the risks associated with change, preparing to deal with the uncertainties inherent in change, and having the facility to integrate specific changes within a longer-term strategic plan are paramount to the success of a manufacturing company. In the best of circumstances, risks can stall growth if they are incorrectly managed; in the worst case, they can shut down a factory.
RISK AS A CONSEQUENCE OF CHANGE
Change and its accompanying risk can occur in numerous dimensions of the manufacturing operation. Changes to the structure of the enterprise, changes in the people in the various roles of the corporation, and changes to the spirit or attitude of management create organizational risks. On the other hand, changes to manufacturing processes, substitution or selection of new component materials, or increased product performance characteristics at the behest of customers expose the firm to technological risks.
When considering change, the arguments against it can be very persuasive, especially when current performance is successful and competitive pressures are unforeseen. Also, changes to complex systems are more likely to be riddled with unforeseen reactions and unexpected risks. But, for manufacturers that intend to drive markets rather than just react to competitive pressures, the active pursuit of change in product technologies and process capabilities must be recognized as an integral component of long-term strategy. The challenge is to minimize the risk assumed and maximize the benefits gained.
While instances of change seem to be easily classified as organizational or technological, in practice they are not easily separated. Organizational and technological changes are intertwined such that successful change in one category often requires accompanying changes in the other. However, we believe that there are ways to increase the likelihood of successful implementation of change and management of risk within our long-term competitive strategy. These include the effective use of people, the adequate planning for change, and providing of the appropriate tools.
EFFECTIVE USE OF PEOPLE
Total Employee Involvement
For senior managers to sit in their offices and assume that they have all the knowledge and experience needed to make critical risk decisions is a prescription for failure. To make the best decisions with the most knowledge, managers have to use the combined intelligence of the entire work force. But just saying this does not make it happen; to capitalize on the human resource and experience requires empowering the work force. Participation in the strategic planning process ensures that all employees understand the direction in which the organization is moving and why. Everyone in the organization needs to understand his or her role in achieving the factory's goals. And when managed correctly and reinforced regularly, people will actively participate in the execution of the plan (see also Badore, this volume).
Most companies that have made effective use of their people have done so using teams. Organizing teams of people into functional or strategic groups is a good start. All people in the teams need training in specific areas to help ensure that the team process will work. These areas include how to conduct and participate in meetings, the fundamentals of project management, problem-solving methodology, and conflict resolution.
As teams take on more responsibilities and assume ownership of their areas, the demand for engineering support declines. This becomes an opportunity for reducing the expense of indirect headcount or redeploying these people into tasks that add greater value.
To be able to reduce indirect headcount, individuals and teams on the factory floor must be able to measure the performance of their area and, when problems arise they must be able to diagnose and correct them. Using the concept of continuous improvement, the bulk of the improvement effort in the organization needs to be centered on reducing variability around the process targets. Although manufacturers have been trained and conditioned to “keep things in spec,” this is not good enough. Successful global competitors continuously try to tighten the process distribution around their targets, which in most cases are well within the “spec limits.” This approach gives each production area a continuously improving process capability. As variability improves and distributions tighten up, there are fewer excursions and thus a need for fewer people to be involved (which will also reduce variability). Using basic statistical process control rules, we can identify which steps are stable and which steps are out of control. These areas are where the technical talent and key teams can have the biggest impact.
Senior management needs to give teams the responsibility and authority to run their areas in both a tactical and a strategic manner. Senior management's role is to understand what the obstacles are and to remove them. Anything less will not build the foundation needed in the organization to deal with change and risk taking.
PLANNING FOR CHANGE
Strategic planning is used to create a “planned crisis” that focuses the people and energy on tough goals that are derived from market knowledge and technological opportunities. At Intel we use our manufacturing capabilities as a strategic market force to drive down the cost and increase yields to prolong the life of each of our integrated circuit manufacturing
facilities. Because of the rapid changes in the technology of the product and the accompanying production processes, we must constantly strive to reduce our costs and raise our production yield. These planned crises provide the numbers and projections to illustrate the importance of increasing yield while squeezing costs.
Strategic planning is the cornerstone of success in the long run. It is a process that must be iterative and gets as much participation as possible. Setting specific goals and brainstorming on how to get there is the start. The senior management of a manufacturing company needs to understand the ramifications associated with reaching the mature phase in the life cycle of a plant or a manufacturing technology, especially in light of more fierce global competition. In addition it is essential to integrate contingencies to deal with the external influences that can derail an otherwise sound long-term strategic plan. New products or technologies can significantly shorten the life cycle of the production facility. Other influences that need to be considered include economic fluctuations, higher than expected demand for products, change in management (and corresponding change in management style), and even mergers and acquisitions. During this planning process, senior management must continue to ask “What can go wrong?” How would we deal with all of these changes, and how would they affect the assumptions that we have used in our strategic model? How would they affect our execution plan?
Clearly, effective long-term planning requires regular feedback of real changes in the environment, the capability of the factory and even new ideas that come from the organization. The long-term goal seldom changes, but details in executing the plan to reach the goal often will change.
Managing Organizations in Transition
Historically factories have gone through major changes in their business or organization. These decisions are almost always made by senior managers who expect people to get on with their new organization, business, job, or supervisor. Yet we are always amazed at how individuals and teams who worked well before a change slide in their performance afterwards. Management usually fails to consider the human emotions involved in organizational change. Writing on individuals going through change and corporations managing transitions, Bill Bridges (1988) describes three phases people go through during changes or major transitions: an ending, a neutral zone, and a new beginning. His work also describes how this process needs to be allowed to happen for a change to come to fruition. Coworkers and management need to help facilitate this process and to be understanding of people during these transitions. We have to plan for change and to with-
hold our normal expectations of people or the organization until they get through the emotional transition.
Everyone, to some extent, will go through this transition process, whether it involves a merger, a major reorganization, being assigned to a new team, or even a plant closure. The difficulty is that individual behaviors and reactions differ. Senior managers of manufacturing plants have to consider the changes and risks associated with long-term competition and profitability. They have to plan for the emotional transitions that the work force, on which they depend, will go through. They have to demonstrate leadership and plan their steps.
PROVIDING PROPER TOOLS
Statistics and Problem Solving
Teams organized to run and improve their areas in a manufacturing operation must acquire the appropriate capabilities in statistical methods and problem-solving techniques. Developing a working knowledge of these methods is essential in monitoring the performance of an area and being able to identify the parameters that afford the highest leverage for improvement. As this capability evolves, there is a corresponding opportunity to implement a process control system used and maintained by production teams and individuals.
The benefits of using such a system are that it promotes teamwork, systematically empowers direct labor production people, improves the learning rate, and improves material quality and product yields. Communications between departments and across work shifts is drastically improved with the use of a common process control system. Properly set up, such a system shares knowledge through documentation. It provides a logical and consistent game plan to deal with variability in the production process and equipment performance. Holes in the existing knowledge quickly become apparent. Engineers spend more time improving the production processes and less time doing routine tasks (such as fixing equipment or adjusting machines to get them back in tolerance).
At Intel we have a mechanism for transferring responsibility for management and control of the production process to the operators of equipment and from the process engineers. This mechanism has supported our strategy to force the costs of the production process down the cost curves. A side benefit, in many instances, has been to stabilize the production process. Our mechanism is a combination of statistical process control (SPC) charts that are used to manage each production process step and response flow checklists (RFCs) to troubleshoot and correct problems. SPC
is applied in the usual manner to data collected and evaluated by the equipment operators. The RFC is Intel's way of transferring, to the operators of the equipment, the knowledge and experience accumulated by the process and equipment engineers as they installed and debugged the production machinery. The information in the RFCs is a series of structured “if then. . . else. . .” statements that lead an operator through a well-defined sequence of diagnostic questions and associated corrective actions or adjustments.
Change Control Methodology
As people in the factory see opportunities to make improvements, there will be increasing pressure to implement changes. In an integrated production process, seemingly harmless improvements can throw a downstream processing step totally out of control. Understanding, through well-designed experiments and characterization, of the interdependencies between processing steps is critical to minimize risk in the improvement process. To help manage this kind of effort, it is useful to organize a change control system. This system should use a structured approach to experimental design, agreed upon criteria for interpreting results, and the confidence level required to implement a change. Customer 's issues need to be considered. When changes in the process or product are considered major, qualification is usually required from the customer.
To maintain consistency with a structured change control procedure requires educating the work force in the philosophy of change control and the mechanics of the process. The committee of people who review these proposals and reports need to represent most of the disciplines in the factory and should be senior managers. The membership of these committees need to be consistent. It is important to ensure that this kind of a system is perceived not as an obstacle to change but rather as a methodology for assessing risk and making decisions in a consistent manner.
In an Intel study of all changes intended to improve factory yields, only 18 percent actually resulted in a positive yield improvement. 1 Yet, in each Intel factory there were change control systems similar to the structure previously discussed. Although many changes were intended to improve manufacturability, the data suggest that the number of interdependencies in the process steps were not well enough understood. We discovered that in a high-yielding, complex manufacturing process for semiconductors, there are no home runs or major breakthrough changes. The biggest gain comes in smaller continuous improvements.
In the manufacture of integrated circuits, yield is measured at several points during the production process: the line yield reflects the scrap rate for entire wafers, while yield-to-die measures the number of good die obtained from each silicon wafer.
Many people express the concern that manufacturing in this country must change, or else our ability to compete and influence industrial and technological growth will disappear. One of the cultural and social advantages that American people have is the ability to take emotional risks in our organizations, absorb changes in a fluid and nondisruptive manner, and deal with evolving group structures and relationships. For manufacturing this is an asset that may indeed be a globally competitive advantage for the United States.
Our ability to capitalize on this asset, however, will require us to develop an excellent working knowledge of change management and the risks associated with it. We must fundamentally break the paradigms established by our education and the structural norms of classical American manufacturing, which have conditioned, and now impede, our competitive opportunities. Senior management must learn how to empower their work force, and they must begin to assume the role of leaders. Employees must be given the knowledge and training to be effective team players and management must establish a culture to promote this type of organizational behavior. As the work force is empowered to take ownership of its areas of the production process, it must also have the tools to understand and improve factory performance. To manage the changes associated with improvement requires an established methodology to review proposed changes, establish the criteria for success, and assess the attendant risks. It is essential that everyone understand and support a long-term goal to be competitive, even if it is not at all clear how to get there. Establishing a strategic planning process will help build a foundation for the factory. This planning process must be iterative and open to all ideas, and it must leverage the strengths and capabilities of the people, plant, and production process.
Once the direction is set and the systems are in place to allow a factory to design change, take risk, and position itself to be competitive, it is critical to establish an environment that reinforces the notion that it is good to take risks. Consistency in management's attitude toward change and evolution of the organization to deal with the requirements is paramount.