-
- News
- Books
Featured Books
- smt007 Magazine
Latest Issues
Current IssueComing to Terms With AI
In this issue, we examine the profound effect artificial intelligence and machine learning are having on manufacturing and business processes. We follow technology, innovation, and money as automation becomes the new key indicator of growth in our industry.
Box Build
One trend is to add box build and final assembly to your product offering. In this issue, we explore the opportunities and risks of adding system assembly to your service portfolio.
IPC APEX EXPO 2024 Pre-show
This month’s issue devotes its pages to a comprehensive preview of the IPC APEX EXPO 2024 event. Whether your role is technical or business, if you're new-to-the-industry or seasoned veteran, you'll find value throughout this program.
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - smt007 Magazine
The Key to Process Control
December 31, 1969 |Estimated reading time: 8 minutes
By Dan Kauss
In the mid-1990s, the 6-sigma quality and defect-free initiative was in full force among the major high-tech manufacturers. With this standard, manufacturers had to look at addressing process inefficiencies and the needs of quality control, which largely were directed toward a search for process monitoring tools that could streamline machine function. The impetus: Manufacturers could not keep up with market demand, and the competition within the marketplace generally would not allow for production delays. Additionally, no third-party vendor of such tools existed, so it forced processors to develop internal solutions. For example, at one company* the goal was to develop a software solution that would track and chart statistical process control (SPC) data in real-time.
Combination of Modules
The development process for the new software began by combining the SPC module with a utilization module. This allowed for the acquisition of sufficient process information to verify quality and improve use, which translated into cost reduction. The tool provides operators, engineers and managers "visibility" into factory performance for improved process control.
The SPC module is required to chart critical processing variables in which the software would offer various measurement- and attribute-charting types. Additionally, it is necessary that the sampling and statistical-update mechanisms, plus the run-rule types, be configurable according to specific needs or user preferences. In short, the software had to identify out-of-control conditions quickly so that root causes could be identified and eliminated.
One major challenge in that solution was the lack of generic communication protocols to enable process-data acquisition. The only way to gain the SPC information from all machines was to use a generic communication protocol.
Existing Protocol Problems
The Semiconductor Equipment Communi-cation Standard/Generic Equipment Model (SECS/GEM) protocol was deemed the best solution for communication from the tool to the host (Figure). Although common in the wafer fabrication industry, SECS/GEM was only beginning to gain visibility within the printed circuit board (PCB) assembly community. Nevertheless, SECS/GEM presented itself as the solution for the proprietary communication issue, which immediately burdened engineers with two problems:
- A software vendor or computer integrated manufacturing (CIM) vendor had to be identified who could provide the SECS/ GEM interface with the unique features required by the PCB assembly industry.
- The latter was proving to be slow in adopting SECS/GEM as the de facto communications protocol.
Several software and CIM vendors were contacted to determine their interest in developing a SECS/GEM interface to be used with the equipment monitoring software. A motion-control and communications-software provider headquartered in Salt Lake City** agreed to begin such a task, i.e., an interface software that would permit information to flow between a surface mount assembly machine and a host-level system. Eventually, this was achieved by developing two SECS/GEM communications interfaces: one residing on the equipment for GEM enablement, and the other incorporated into the equipment-monitoring software.
The GEM Host Manager - a plug-and-play SECS/GEM equipment interface - was the result of the collaborative effort, providing a complete communication package that could be incorporated into the process-monitoring software tool.
A Single Software Development
It soon became clear that the PCB industry would be served best by a single software application, one that would use SEMI standard E10, which defines six equipment states: Productive, standby, engineering, scheduled down, unscheduled down and nonscheduled. Such a single software tool would increase production capacity using existing capital equipment.
Monitoring software development began by documenting the specific requirements of an assembly-monitoring tool that allowed for real-time coverage. Initial findings showed that many factories still were using internal and independently developed applications that provided some monitoring functionality but did not meet the specific goals for yield enhancement, line efficiency and scrap reduction. The Kempner-Tregoe decision-analysis process eventually was selected to decipher the information gathered.
Once data were collected and evaluated, it was determined that there were four key modules to incorporate into the new software: SPC, alarm-management, utilization and Web enabling. Individual consideration illustrates the value each plays in overall benefits of the software tool.
- The SPC module addresses the cost of poor quality by providing real-time statistical information in easy-to-read charts and graphs that offer time-, event- and unit-based samplings as standard features. Additionally, the user can configure charts as needed to provide the analysis required to maintain constant production quality. By maintaining an established level of quality, such as 6-sigma, the factory can minimize the cost associated with analysis, repair and retesting of defective units.
Within the SPC module, Western Electric rules define SPC violations or out-of-control events. It is important to give the user the ability to configure certain SPC parameters including actual values vs. standardized "normals," short run vs. long run, sample size, and frequency and control limit calculations. By establishing parameters tailored to the needs of the factory and configuring control charts in an easy-to-read and interpret format, the factory has the tools required to maintain quality standards.
- Events are monitored as they occur in the SPC charts and are based on the defined control limits (either set manually by the user or automatically by the system). It will notify operators, engineers and other specified personnel of out-of-control events in real-time. This is part of the Alarm-management module.
Notification of alarm events via client, pager and e-mail allows for immediate response and problem correction. The system module offers on-line out-of-control action plans (OCAP) that guide equipment operators automatically through key steps for issue resolution. OCAPs are integrated into the operator chart view automatically; its procedure is presented on the client screen to the operators during out-of-control conditions. Operators then must log in and choose the corrective action from the OCAP list to clear the alarm. At the same time, the process-monitoring software logs the alarm, time, machine, operator name and OCAP action, all of which simplify the tracking and enforcement of out-of-control event resolution.
The third aspect of process monitoring, Utilization, provides actual pure data for each machine's performance. Data are collected in real-time and displayed on the client screen to visualize multiple metrics and to identify where lost opportunities can be leveraged to improve asset use. This module can identify bottlenecks, determine optimum build rate, calculate mean cycle time and conduct line-balancing analysis. The final module to be addressed is Web Enabling, or user access to historical data through a company's intranet. It permits data to be shared throughout an organization with hard-process data made available to multiple factories. The software functions as a parallel collaboration tool, with which plant managers can see how machines and lines in other facilities are performing.At this point, an education system comes into play. As one factory finds a solution for a quality or productivity issue, the information can be shared with other factories experiencing the same or similar problems. Techniques can be communicated to benefit each production facility, thus benefiting the entire corporation.
A "Real-world" Test
Once the application requirements, software development and GEM Host license were combined into one package, it was necessary to determine how to test and verify software accuracy. Preliminary testing was conducted in four factories, two domestic and two off-shore, while engineers worked on verifying the data collected and applying the results to improve productivity, reduce scrap and increase process control. The testing period also provided time to implement new features and user interface improvements.
The first "real-world" test of the software was conducted on a line manufacturing new cell phones that had been unable to meet market demand. The factory had to increase production without purchasing additional capital equipment. The new equipment-monitoring software was installed on the line to collect machine-performance data in real-time. After analysis, improvement opportunities were identified that resulted in a 12 percent production capacity increase.
A second study then was conducted to test the real-time monitoring capability of feeder/nozzle misplacements and to fulfill the quality system review (QSR) requirement of process control for placement equipment. Control charts were created to pinpoint out-of-control conditions and to supplement detailed Pareto charts. Using the monitoring software, the factory reduced "mispick" rates by 30 percent and process variations by 50 percent. With these results the software passed ISO 9000 qualification for the QSR process-control requirements.
One major challenge in implementing process-monitoring software was (and continues to be) convincing engineers of the tool's viability. This is because process data traditionally have been collected manually. But with the introduction of process-monitoring software, all data collection would be automated. Hence, the necessity of maintaining some level of security in the numbers reported is removed, and the data collected are pure and comparable between factories. While a culture change, the benefits of these types of data are immensely valuable to the manufacturing enterprise. Actual process-control data now are available throughout the corporation rather than risking quality default via estimated calculations and assumptions.
Verifying Viability
To demonstrate the process-monitoring software's viability, it was determined that there must be a 2 percent use improvement. The results of the implementation and utilization, however, increased production capacity from 2.2 to 8.7 percent. This result was demonstrated in a factory that realized a 38 percent capacity improvement based upon data and feedback from the new monitoring software.
The feedback provided by the software permitted the factories involved with the beta studies to monitor equipment performance. With insight into each piece of equipment, bottlenecks, nozzle-vacuum degradation and feeder shortages were identified. To further demonstrate the viability of the new tool, one factory benchmarked the losses incurred weekly on piece scrappage, an average charge of $2,270 per machine per week. Monitoring the line, identifying quality improvement opportunities and implementing efficiency changes reduced piece scrap loss to $490 per machine per week.
Conclusion
As high-tech manufacturing moves forward, it is essential that the assembly process continues making strides in efficiency and quality to maintain cost competitiveness. By using process-monitoring software, factories can identify inefficiencies and quality variations in real-time - the "difference" in enabling process engineers to make the necessary changes to maximize machine function, increase yields and reduce costs.
*Motorola Inc.**Cimetrix Inc. GEM Host Manager is trademarked to Cimetrix Inc.
Dan Kauss is lead applications engineer at Motorola Inc., Global Software Group, 50 Northwest Point Rd., Elk Grove Village, IL 60007; (847) 907-8662; Fax: (847) 907-7774; E-mail: www.dankauss@motorola.com; Web site: www.motorola.com/manufacturingsoftware.com.
Figure. The SECS/GEM protocol, used for process control by the wafer fabrication industry, was used as the basis for development of a process-monitoring software tool for PCB assembly.
null