No productivity gains without IT planning
01/01/1999
FAB MANAGEMENT
No productivity gains without IT planning
Scott Mullarkey, EDS Hi-Tech SBU, Phoenix, Arizona
Productivity is defined as the rate at which goods or services are produced. Most semiconductor companies, however, have expanded this definition to include concepts of yield, scrap, inventory turns, and cycle time. Once the importance of the refined definition is understood, information technology (IT)-based systems are put in place to support the collection of data, so that meaning can be applied to the concepts. Feedback loops are added so that improvements can be based on information analysis. These processes sound good, but why don`t they work consistently?
Even if a common definition of productivity exists within an enterprise, delivering systems to support that definition often seems to take forever without ever quite doing the job intended. Different agendas and seemingly disparate requirements usually spell doom for such attempts. We are all familiar with islands of automation - information systems that do not meet changing requirements or that are simply not used. What can we do differently to prevent these problems? Are there common threads to the success and failure of such efforts?
Start at the top
The typical journey of IT-based systems follows business intent. As the figure shows, the state of IT-related systems over time - those originally put in place to support the mission of the enterprise ("IT capability") - becomes misaligned with the direction of support systems. As the gap between business intent and IT capability increases, we do not add value to the new way of more thinking, and we create an environment where IT impacts the enterprise negatively. A "quick fix" can add strength to IT capability (in effect, thickening the blue line in the figure), but this only helps to prevent a meltdown; it does not provide added value to support productivity gains. The fundamental error assumes that IT capability magically follows business intent, an expectation commonly propagated within an enterprise. It takes more than magic to make that happen.
 |
Typically, as business focus and intent change over time, they become misaligned with the direction of IT support systems.
Changing a company`s strategic intent can take months, even years. As fundamental as it sounds, the insight and vision start with the CEO. He or she has the prime responsibility of charting the course for the enterprise. The only true way to bring added value to the corporation is to understand where the company is headed and provide ways of supporting the journey.
Consider, for example, moving from wire-bond interconnect to bumping or chip-scale packaging, which is increasingly showing the potential to impact semiconductor manufacturing productivity dramatically. In the definition process, critical success factors become apparent and are the subject of focus. These factors include quality, manufacturability, design-time turnaround, market penetration, and customer support. In a broad sense, the definition of acceptable productivity is outlined within a business strategy that will move any new concept forward; the business plan makes certain assumptions about the definition of success in making any new concept viable. These parameters must be carried into a business planning and implementation process that includes IT.
At this point in the business-planning process, discussions must include how IT can be used and must evolve to support critical success factors. A key question is: Are current systems for collecting performance data adequate to provide the timely feedback necessary to support business goals? Close analysis of answers to this question redefines productivity to include the ability to support business goals; the meaning of productivity must also be expanded to include:
internal and external supply-chain management performance (delivery, quality, schedule);
capital investment and financial planning (return on investment, cost models);
operational performance (yield, cycle time, scrap, turns, schedule performance);
sales and marketing performance (market penetration, growth, new markets);
human performance (turnover, training, efficiency); and
research and development performance.
Timing is critical. Without defining and planning for IT impact with these factors in mind, it is not reasonable to expect to meet even conservative estimates of system design and implementation to support basic aspects of critical success factors. Simply installing a piece of software or integrating systems will not solve the problem. The implications are immense. Virtually every system in the enterprise will be affected, including enterprise resource planning (ERP), manufacturing execution systems (MES), computer-integrated manufacturing, human resources, planning, and the supply chain.
Where to consider IT
As business systems become more complex and far-reaching, making fundamental changes in doing business causes fundamental changes within IT-based systems. To accomplish these changes effectively, IT must be involved with the development of the business plan; solid IT architecture is required to accommodate change and new technology; a well-defined systems engineering methodology must exist; and a well-defined project management methodology must exist.
These elements are required to identify and plan for change, then design for and implement change. Not applying these concepts will invariably create expectations that are not realistic in the development and implementation of systems that support productivity goals. The real-life effect is the inability of operations and engineering departments to collect, analyze, and react properly to data in a wafer fab. This can result in higher scrap, lower yield, more misprocessing of wafers, and lower throughput. A large part of productivity is our ability to react to changing requirements quickly.
A practical example
The recent start-up of a semiconductor manufacturing wafer-bumping operation demonstrates how the process of planning with IT in mind works. Understanding the critical success factors within the business and mapping IT capability was a key to success. Critical success factors were developed by the executive staff to include:
world-class wafer bumping, with specific targets designated by the CEO;
shortest possible turnaround time of customer wafers, which is important because wafer-bumping service is an extension of a customer`s manufacturing process;
low scrap, which is important when processing someone else`s wafers;
cost efficiency, because bumping competes with mature, low-cost methods; and
effective management of a continuing stream of new products and differing design and manufacturing requirements based on wafers supplied by the customer.
These factors became the driving force for making IT-related decisions. In this case, a greenfield manufacturing site was established. This dictated that certain actions (decidedly different from those taken in a legacy manufacturing situation) be taken to support the goals of the enterprise:
An IT roadmap should be developed that outlines basic technology choices to be supported. This allowed more efficiency with limited resources. Instead of letting application or design choices dictate the IT, we chose to deploy a "least common denominator" of technology to be the basis for application, development, or integration choices. The roadmap included direction on such items as hardware platform, networking protocol, relational database, e-mail, and operating systems. These choices limited what was available to us from a technology and product standpoint, but the benefit was that we are now better able to support the enterprise goal of building in cost and time efficiencies.
A MES architecture should be developed that supports the ability to set up and manage new products quickly, and to collect and feedback appropriate information to support engineers and technicians in quickly tightening the quality loop. Most important, design decisions that supported the enterprise goal of quick turnaround and efficiencies were made with the knowledge that this goal was closely coupled with the future ERP and planning and supply chain management systems. Regardless of the products or technologies used, integration efforts must facilitate timely information proliferation to support the enterprise.
A closely coupled team of operations, engineering, management, and IT expertise should be established to manage the complexity of new developments and integrate and support the MES. This includes rigorous project management and systems life-cycle methodologies not only to set the proper expectations about what can be accomplished in a certain time frame, but to deliver software and support services that are world-class in themselves.
These fundamental elements become the basis for how systems are designed and deployed, resulting in the ability to measure IT`s effectiveness in meeting the goals of the enterprise. The trick now is to continue this process by understanding that new ideas are emerging about how to run the business; long-range planning will dictate what those changes are and how IT should be used to support them.
As these issues become more prevalent in the semiconductor business, look for a small number of IT providers to develop the capability of supplying top-to-bottom systems in support of increasingly difficult business propositions such as 300-mm fabs. This will include the integration of tools in days, not months; scaleable systems architectures to meet quickly changing needs; comprehensive consulting capability to solve business issues with IT; and the ability to become a true business partner by applying the vision of company leaders to the task at hand.
The definition of productivity has expanded beyond "the rate at which goods or services are produced." The new definition should read something like: "producing goods or services at a higher quality, a lower price, and in a shorter time than yesterday for a set of business requirements that will be different tomorrow." The consequence of not "producing" is to be out of business more quickly than you can imagine.
SCOTT MULLARKEY is an account director in the High-Tech Semiconductor Business Unit (SBU) of EDS, 3701 E. University Drive, Phoenix, AZ 85034; ph 602/431-6020, fax 602/470-0360, e-mail [email protected].