Like scientists, plant managers are identifying and testing hypotheses about every part of the manufacturing process, from warehousing, power consumption, raw material waste and product quality to efficiency and production rates. The goal is to identify small improvements—1 percent to 2 percent—in each part of the process that add up to significant—20 percent to 30 percent—improvements overall.Part of the challenge for manufacturers is that their laboratory is a high-speed, high-production environment, with strict ergonomic and safety standards. In this environment, managers must use technology to its fullest advantage.The majority of manufacturing facilities and processes are already being controlled by some sort of processor-based system. These controllers have been around for decades, competing with one another to become the accepted standard. While no controller is the declared winner, the competition has led to many advances in industrial automation, including establishing standards, industrial networking, embedded Web pages, increased speed and capabilities—and phenomenally low prices.One thing controllers have in common is that they support a Microsoft Windows interface. Now you have the ability to consolidate data from multiple sources on a single, common platform. Regardless of your current controllers, you can pull data from them, put it on a server and consolidate it with data from other personal computers (PCs), controllers or even your keypad. You have the tools to continuously interrogate your process equipment and monitor its performance, without having someone sitting there day in and day out dedicated to it.For example, using a data collection system of this nature, you can measure the efficiency of process equipment against a known/fixed baseline. This allows you to identify bottlenecks and target projects to eliminate them. Eliminating your prime bottleneck has a direct, net positive effect on your productivity. In addition, you get the return on investment (ROI) information required to base capital expenditures on quantitative data, rather than instinct.Preventive maintenance is another area that benefits from data collection systems. All equipment has a mean time to failure. Key areas to monitor include motor run time, motor starts/stops, stoke counts, operating current and vibration levels. By identifying the sources of equipment downtime and predicting failures, you can schedule preventive maintenance during a planned downtime.
Improve qualityData collection systems also allow you to improve quality. By correlating the quality of product with your operational variables, you can identify causes of out-of-spec product and focus your efforts on eliminating the variability that caused it. Related to quality assurance is the ability to trace product and track inventory. By correlating resources/vendors with each product or batch, you can develop a complete product lineage for product recalls, complaints or praise, and identify the responsible batch, vendor, resource, operator and/or customers affected.Performance tracking can be improved, as well. By correlating each operator with important performance indicators, you can track how they do compared to a baseline, or to other operators. This quantitative data can then be used to motivate and challenge operators to improve overall production and efficiency.The final area ripe for incremental improvement is resource planning. Based on the product/batch being run and the historical cycle time for that product/batch, you can allocate inventory, equipment and people required. As new orders come in, you can verify resource availability, experiment with different scheduling and optimize your process.Every plant manager knows you can’t predict everything, but the ability to collect, store and correlate data makes a science out of what was once an art.
Michael Gurney,
[email protected], is a partner in Concept Systems Inc., a Certified Control System Integrator, based in Portland, Ore.
Subscribe to Automation World's RSS Feeds for Columns & Departments
