Click to learn more about author Boris Shiklo.
In this piece, we’ll explore how IIoT drives productivity in the plastics industry and provide an example of implementing an IIoT solution for monitoring equipment effectiveness at a plastic parts manufacturing enterprise.
Volatile raw material prices, the lack of skilled labor, and the complexity of the global supply chain — these and other issues contribute to a decrease in the plastics segment’s efficiency. It has been reported that over the past year the industry’s productivity has declined by 3.3 percent. To change the situation for the better, a number of plastics manufacturers have turned to modern tech and have started adopting the Industrial Internet of Things (IIoT) for monitoring and analyzing shop floor performance.
As the experience of early adopters can be helpful for the manufacturers who are at the start of their IIoT journeys, in this article, we will share an example from our IoT consulting practice and show how a U.S.-based plastics manufacturer leveraged IIoT to tackle the issue of insufficient equipment effectiveness. We will also examine the benefits they’ve achieved and the struggles they’ve encountered on the way to a connected shop floor.
The IIoT-Driven Approach to Monitoring Equipment Performance
With IIoT in place, manufacturers get the opportunity to view equipment effectiveness metrics (e.g., runtime, downtime, cycle time, the number of parts produced) in real-time without physical access to machines. For that, the data about equipment’s operational parameters (e.g., a machine reset signal, pen drop, pen lift) is fetched automatically from machines’ PLCs via serial or Ethernet interfaces.
Equipment’s operational data is relayed over to the cloud software — the core of an IIoT solution — for storing and analysis. The analytics module of the cloud software turns equipment’s operational parameters into informative insights about machines’ availability and performance (e.g., uptime, downtime, cycle time, total parts count, etc.). The insights obtained with the analysis are visualized and presented to the plant’s staff upon request via web or mobile applications.
The Implementation Example
To illustrate how the described approach works in practice, we’ll discuss an example of a plastics parts manufacturer who has leveraged IIoT to monitor shop floor performance in three geographically distributed manufacturing facilities across two manufacturing divisions: die-cutting and machined plastics. The manufacturer rolled out an IIoT solution to address the issue of unreliable machine performance data arising from manual data collection. Below, we describe the implemented IIoT solution from the perspectives of machine connectivity, data analytics, and communication with the users.
In both manufacturing divisions, the enterprise uses machinery with computer numerical control (CNC). At the machined plastics division, the enterprise employs CNC milling machines, CNC lathe machines, and CNC routers. At the die-cutting division, the manufacturer uses hydraulic, traveling head, beam cutting, and receding head presses, all with computer numerical control. The equipment comprises both legacy and newer machines, which have different connectivity interfaces. This influences the way in which the machines are connected to the IIoT solution.
- For legacy equipment lacking Ethernet connectivity, the operational data is sent from a machine’s PLC through a serial port (RS-232, RS-422, RS-485) whenever there’s cycle start/end, spindle on/off, etc. The serial port is connected to the serial-to-LAN converter, which forwards the data to a cloud platform via an IoT gateway.
- For CNC machines supporting Ethernet communication, the operational data is sent from the machines’ PLCs to the cloud software through an Ethernet port via an IoT gateway over a wireless network
The enterprise wanted to conveniently see equipment availability and performance metrics (e.g., operating time, downtime, total parts count, etc.) across the machined plastics and the die-cutting division for varied timeframes and to be able to compare machine effectiveness across production lines and factories. Along with it, the manufacturer wanted to be informed about possible critical production issues, such as machine failures, as soon as they arise.
To meet the requirements of the enterprise, the IIoT solution was geared with two types of data analytics: batch and near real-time.
With batch analysis, the operational data collected from a machine’s PLC via a serial or an Ethernet interface is transmitted to a cloud data storage, where it is aggregated for an appropriate period of time (e.g., hour, shift, day, etc.) before it is analyzed. For instance, the operational parameters of a CNC milling machine are aggregated for an 8-hour shift and analyzed at the end of it to generate an equipment effectiveness report per shift, featuring such metrics as total power-on time, total motion time, average cycle time, and the total number of tool changes.
Near real-time analytics implies collecting and analyzing equipment data immediately after it is generated. Near real-time data analytics is used to provide a fast output in the form of an alarm informing responsible staff of potentially critical situations — for instance, a traveling head press abruptly stopping during an operation cycle.
Communication with Users
The insights obtained with the analysis are communicated to the shop floor and factory managers via web and mobile applications. The output the solution provides takes the form of reports and near real-time alerts.
The solution allows factory and enterprise managers to generate basic and extended reports. Basic reports display the data about CNC machines’ statuses, uptime, downtime, the number and the duration of cycles, and other availability and performance metrics for a selected period on a dashboard.
The extended reports provide the ability to explore an entire machine’s event history for any period and see every machine’s effectiveness dynamics. For instance, a COO can view a monthly availability and performance report for the die-cutting division and compare the obtained metrics with those of the previous month, the dynamics conveniently visualized in the form of a line chart.
Alerts are generated when equipment data shows patterns critical for the cutting or pressing operations. The equipment operational data is analyzed in near real-time against the rules defined by data analysts in cooperation with manufacturers. The rules determine potentially critical situations and respective actions that should be taken whenever such a situation arises. For instance, if a CNC milling machine rejects the start of a new cycle, an alert is triggered and sent to a maintenance specialist and a shop floor manager via a web or mobile application.
The Benefits Gained with IIoT
The solution has driven substantial operational improvements across the enterprise, the most important of which include:
- Instant Access to Shop Floor Data: Due to the improved collection, aggregation, and processing of data, equipment effectiveness reports become available to the enterprise and shop floor managers in a matter of minutes, so that they are always supplied with accurate equipment effectiveness KPIs.
- Precise Visibility into the Shop Floor Operations: At the shop floor level, IIoT has provided shop floor managers with the ability to see the current level of equipment availability and performance, as well as get notified about potentially critical situations in near real-time. At the enterprise level, the implementation of IIoT has given enterprise managers visibility into how well each manufacturing division is performing and provided an opportunity to see performance dynamics across divisions.
- Detailed Analytics: The insights obtained with IIoT go well beyond the information about a specific machine’s uptime and downtime. IIoT allows you to see detailed performance metrics for each machine, production line, factory, and manufacturing division. For instance, for CNC routers, it is now possible to obtain the data about the average cycle time, the last cycle time, the number of tool changes, the number of tools in use, the number of parts produced, and more. This has enabled enterprise managers to leave behind the guessing game and start making data-driven decisions about improving equipment effectiveness across the enterprise.
- The Ability to Monitor Labor Effectiveness: By correlating equipment effectiveness data with the data about operators working on particular machines, it is now possible to get insights into the performance of each operator and get a better understanding of the overall labor effectiveness. A shift manager can, for instance, see that during a shift, John operating CNC lathe X has produced 12 parts more than Jim operating CNC lathe Y has.
The Challenge on the Way to a Connected Shop Floor
A challenge that substantially complicated rolling out an IIoT solution for this manufacturer was the need to connect legacy equipment to the cloud software. Some of the enterprise’s hydraulic and receding head presses were designed several decades ago, so they do not support modern connectivity methods. To make the presses communicate with the cloud, it was decided to equip them with sensors, which in their turn, were connected to an external controller. The sensors would gather the operational metrics (e.g., press power on/off, table lift, etc.) and pass them over to the cloud software through an external controller via an IoT gateway.
Outcomes and Future Plans
The automatic collection and analysis of equipment data allowed this manufacturer to tackle the issues they had struggled with when collecting and analyzing equipment utilization data with manual methods — data inconsistency and its availability with delays. The implementation of an IIoT solution for monitoring equipment effectiveness allowed the enterprise to see how well the machines were performing across both die-cutting and machined plastics division. The enterprise has since experienced considerable operational improvements and plans to extend the solution’s functional scope to be able to track yield quality.