Transforming manufacturing operations with machine data is now a reality for electronics manufacturers and those embarking on this journey are finding themselves further and further ahead of the competition.
The Digital Transformation Imperative
Manufacturers need timely, accurate, and actionable data to chart a path for optimizing manufacturing operations. But the lack of rich, precise data gathering and effective analytical processes has held most manufacturers back. Until now. Now, a hyper-specialized tech industry exists to find the right solutions. The ground under our feet is shifting and the manufacturers who leap for industry 4.0 will find themselves head, shoulders, even skyscrapers above their competitors. Transforming manufacturing operations with machine data has become imperative for manufacturers today.
Arch Systems, the industry’s premiere machine data and advanced analytics provider for electronics assembly operations, is the partner who is dropping down the rope to assist that leap. Leading manufacturers around the world have implemented ArchFX solutions to improve their operational efficiency. They gain competitive advantage through Arch’s support for data extraction and transformation, global KPIs, and predictive analytics that provide actionable insights.
The What and the Why of Industry 4.0’s ETLA Process
Any viable company-wide industry 4.0 manufacturing solution must enable the Extract-Transform-Load-Analyze (ETLA) process:
1. Extract data from diverse sources.
2. line managers can easily view and understand KPIs, and
3. manufacturing engineers can optimize machines and configurations to improve productivity.
Here’s where Arch Systems comes in, every step of the way.
Extract Data
To have the complete ETLA-to-optimize pipeline, manufacturers need data from diverse sources and the right tools in place to take action:
Use of existing integration APIs and standards
Industry 4.0 solutions must be able to leverage standards like OPC UA or IPC-CFX—and know how to use other techniques where those standards don’t exist. Arch Systems has deep experience integrating its ArchFX Broker with different protocols and can leverage industry-standard or vendor-specific protocols.
Configurable database connectors
Frequently, manufacturers have useful data—just not all at one line, area, or site level and often stored in old tools that don’t have the benefit of modern APIs for programmatic data extraction. The ArchFX Broker includes database connectors that support commonly used databases out of the box—yes, even the oldest and most arbitrary—and can be easily configured to support proprietary or vendor-specific databases as well.
Pick-and-place (P&P) machine connectors
P&P machines are the beating heart of an SMT assembly line. Any SMT solution for calculating KPIs or doing predictive analytics will need out-of-the-box integrations with leading P&P machines. No problem. For example, on Fuji P&Ps, Arch Systems can extract data using NEXIM, FujiTrax, or Flexa APIs running Host Interface; on ASM P&Ps, we can use both current and past versions of ASM OIB; and on Panasonic P&Ps, Arch uses either PanaCIM® or direct database connectors.
Automatic optical inspection and in-circuit test machine connectors
Automatic optical inspection (AOI) machines perform a critical quality assurance check as boards come off the line. In-circuit test (ICT) machines provide a deeper level of quality assurance testing by verifying that circuits function as intended when plugged into testing boards. From first tests to last, Arch supports both by collecting information and rolling it into session analysis.
Programmable Logic Controller (PLC) data taps
Arch Systems can easily connect to your existing PLCs, gather useful data from them, and, for the first time, make that data available for other applications.
Fieldbus connectors to tap the native internal language of machines
Some modern digital machines lack a database, industrial PC, or an integration API. In such cases, it may be necessary to tap the machine’s internal communication bus, sniff the data packets, and create a fieldbus integration that understands the machine’s native internal language and extracts useful data for the manufacturer. Arch Systems is experienced in developing sophisticated fieldbus integrations that process tens of thousands of internal data packets per second and efficiently extract the data of interest.
Sensors for legacy machines that generate no data
Arch Systems offers the modular IOTile POD framework for instrumenting legacy machines. Off-the-shelf IOTile PODs monitor common metrics of interest such as temperature, humidity, shocks, current, and voltage. The modular architecture makes it easy for Arch to rapidly develop a new IOTile POD or Accessory for as-yet unsupported values, integrating the needed sensor module and quickly manufacturing the new IOTile device. This can all be done at low cost in low volumes for a pilot and in the high volumes needed for production deployment.
Wireless connectivity to quickly link distributed machines on the factory floor
Manufacturing shop floors were often designed in an era before ethernet, let alone the internet. Arch Systems IOTile PODs support industrial-grade Bluetooth® Low Energy beacon-based connectivity so machines in any location—even the noisiest electromagnetic environments—can generate data without requiring ethernet or WiFi. We also know that sometimes it’s not necessary to instrument every machine on the line, so we’re as subtle as we are robust. Arch Systems can tap material-handling systems such as conveyors or assembly lines by tapping common ports or by installing beam breaks or other sensors.
Translate and Standardize Data
Once data has been extracted from diverse machines, it must be translated and standardized to enable the calculation of KPIs and the development of predictive analytics. The ArchFX Broker is a flexible message broker that can accept all data but is smart enough to filter out the useless; can efficiently store all data in cloud data lakes yet also compress the key parts; and can securely transmit findings to make that information useful by everyone from line managers to manufacturing engineers to site, regional, and global executives.
Find the Root Cause
Traditionally, the large amount of available data has itself become a barrier to recognizing which data matters and is actionable. Arch analytics extract the signal from the noise so the signal is understood, prioritized, and becomes actionable. We focus on feeder and nozzle analytics (to keep the P&P machines unclogged); session analytics (to learn from uninterrupted periods of build activity); and centralizing data (so no one has to wade through per-machine error report menus).
Act, Don’t React
Once all the data is gathered and translated, we help you see within it to where the most actionable issues are. For example, in the case of one EMS manufacturer, analysis of 15,000 pick-and-place feeders over one uninterrupted period showed that 33 percent of the mis-picks were caused by just 0.3 percent of the feeders. You can guess how quickly they prevented the next mis-pick from ever happening.
Drive Manufacturing Execution and Procurement
A successful industry 4.0 solution must be able to accommodate multiple manufacturing execution systems (MES) and enterprise resource planning (ERP) systems within one organization. The ArchFX Broker can extract data from MES and ERP systems, send standardized data to them, and securely move data to the cloud for analytics and machine learning.
Learn to Fish; Never Go Hungry
Arch customers are using the data lake, APIs, and diverse business intelligence tools themselves to test hypotheses about site-specific operational problems and opportunities for improvement. In this way, top-level corporate KPIs can be driven from a single source and the same data can be used by all sites in customized ways while still connecting to the advantages of having centralized, standardized data
Manufacturers can now collect complex data from both new and existing machines, automate the collection and standardization of data via the central broker, and achieve real-time analytics results from machines distributed across the world. The power of a central data lake, combined with real-time information, enables manufacturers to act faster to solve problems, optimize assets, and generate increasingly powerful analytics for predictive capabilities. Digitization initiatives can roll forward at an increased pace as they add more machine categories- expanding the ROI of machine utilization while also tapping into the data lake for the next set of advanced use cases.
Like what you see? Read more about transforming manufacturing operations with machine data in this free white paper or watch our recent interview with Global SMT & Packaging’s Trevor Galbraith.