There are costs associated with waiting for inventory, and disruption to the normal flow of operations can cause inefficiencies. Furthermore, commitments to customers may be affected, potentially leading to financial penalties.
On the other hand, it can be important to stop the line if a problem is detected to prevent the continued production of defective units. The operator or production supervisor needs to be notified as soon as possible in case of an error. Line stoppages are undesirable, and manufacturers must ensure they occur as infrequently as possible to guarantee optimal productivity and minimize unwanted costs. When stoppages cannot be avoided, production must be restarted as quickly as possible.
Troubleshooting Tools:
Today's assembly lines are packed with sensors that detect problems such as clogged nozzles or pick-up misalignment. Automated inspection can be performed at multiple points along the line, even after screen printing, component assembly, and reflow, and the results are used to assist with real-time troubleshooting.
Manufacturers rely on software tools to help manage production equipment, detect any exceptions or defects, and provide information to production supervisors. These include dashboard applications, which provide an instant assessment of equipment status and production. There are also value-added tools, such as Yamaha's QA Options and the closely related Mobile Judgement application. Mobile Judgement can send details of any detected defects, including inspection images, directly to the supervisor's mobile phone to help assess the severity of the problem. Based on their judgment, the supervisor can then order the line to stop immediately, if necessary, until the issue is resolved.
The human-machine interface (
HMI) represents the point where machines and humans interact to solve the problems that inevitably arise during the normal course of a shift. The team in charge must apply its skills to deliver the required quantity of each batch and ensure that production proceeds as planned.
This approach may depend on the quality of the tools provided for visualizing the data captured from various locations along the line. Identifying the causes of failures and quickly implementing a solution also depends on the skills and knowledge of the supervisors present at the factory. If the company's main supervisor is unavailable for any reason—perhaps working a different shift, on vacation, or on sick leave—productivity could suffer.
The latest advances in data science can help eliminate reliance on subjective judgment by automatically identifying the root causes of defects or other production problems. Guiding operators directly to these causes leads to more consistent production performance. As an example of this new generation of software tools, Yamaha's latest tool, the Yamaha YSUP Analytics Dashboard, contains advanced tools that use statistical techniques to identify the most likely causes of detected defects. It also recommends appropriate corrective actions. The analytical tools complement the graphical tools that help visualize equipment performance on the production line (Figure 1).
Figure 1. The control panel visualization tools help evaluate process performance.
The new YSUP Analytics Dashboard presents information in an easy-to-read format that can be understood at a glance. Ultimately, this enables all work shifts to deliver consistently high production and end-of-line throughput, and dramatically reduces downtime.
Based on a relational database with built-in query functions that analyze data on-site, the software identifies the likely causes of logged exceptions in near real-time using statistical techniques. The new dashboard application's automated analysis features allow any authorized user to extract the information they need, analyze errors, and view corrective actions at any time.
Available tools include automated pick error analysis, which uses analytical methods such as flow evaluation, error code diagnostics, status check data diagnostics, and image recognition diagnostics. The user can view the time and circumstances of any pick error, see its exact location, and view corrective actions (Figure 2). Yamaha has verified the accuracy of this tool's automated pick error analysis in experiments conducted with a major customer, a leading automotive manufacturer. Built-in features also support root cause detection, including managing the results of automated assessments and tracking the content of corrective actions.
Figure 2. Analysis of picking errors, including location and corrective measures.
The power of the underlying platform allows for the addition of other analytical tools, such as location and print analytics. Furthermore, the platform is open to leveraging artificial intelligence (AI) in future releases, which will provide additional analytical capabilities and further accelerate problem-solving directly on the production lines.
Conclusion:
Automation has enabled electronics manufacturers to eliminate human limitations from many activities on production lines. However, intervention has remained necessary in cases where assessment is required and assistance is needed to manage exceptional circumstances and restore proper operation. The latest data-driven dashboard tools provide additional support in identifying and addressing the root causes of errors, helping to overcome operator knowledge limitations, and shortening restoration times, ultimately leading to further productivity improvements.
