Las Vegas -- Activplant Corp. today said its patent-pending Throughput Analyzer has entered general release, an application that takes manufacturing intelligence one step further into the area of performance management.
The company is so sure of its ability to help manufacturers increase plant performance 5% to 25% by optimizing what already exists, it also offered a guarantee. If the customer does not achieve its throughput improvement goals within 14 weeks -- the typical deployment timeframe -- it has no obligation to pay for the software, which ranges in cost from $100,000 up to $1 million-plus.
Activplant officials made the disclosure at Managing Automation's Progressive Manufacturing Summit, here, as it prepared to release the names of its first customers of Throughput Analyzer. Beta tests have been conducted at a large Japanese automotive OEM and a tier-one automotive supplier. According to the company, by using the software, a $100 million plant can achieve at least a 5% increase in throughput resulting in a $2 million-to-$3 million annual savings.
"Our customers tell us this is really big," said Dennis Cocco, Activplant's founder and chief product strategist. "That's because the technology surpasses typical overall equipment effectiveness (OEE) metrics used to measure plant performance -- an often arbitrary process that can provide statistics on how a line is operating, but can't pinpoint the exact machine or machines that is causing degraded performance, such as downtime or impaired speed.
Throughput Analyzer is able to identify the chronic constraints -- that is, what is impeding the flow of products through the process all of the time. The application gathers just three pieces of data from each PLC on the shop floor. Its patent-pending algorithms then track four critical cycles including, uptime, downtime, speed losses, and quality losses.
That information is captured and funneled into a real-time logic engine to measure every machine, giving a snapshot of how specific machines and production lines were performing against an ideal measure of achieving perfect "Takt time" (a lean concept of measuring the amount of time a process takes). The user can view and drill down into data in a graphical format that makes it easy to pinpoint and address problems.
Cocco recommended customers analyze data over a period of time such as two to three weeks to get a better understanding of chronic constraints rather than anomalies. "That way, they can marshal their resources to solve the real cause of the problem," he explained.
Throughput Analyzer includes a "root cause assignment" module that companies can apply to problematic areas in order to identify exactly what is holding back performance. The system will identify problems in human operators as well as equipment, he added.
What makes Throughput Analyzer different from existing manufacturing intelligence applications is the information. "We are not tapping into data that exists, we are measuring performance and creating the data," said Cocco in an interview.
The technology, introduced in February, is meant to be "simple," Cocco said. There could be upwards of 150 pieces of information coming off of production equipment, but the Activplant algorithm boils it down to three measurement points: waiting for parts; blocked for parts; and reaching the end of the cycle. "We just care about the movement through the cycle," he said.
Page : 12 ... NEXT
 |
|
|
|