Part one of a five-part series on Manufacturing Intelligence (MI). This is a high-level overview intended for all persons involved in the data stream, including: Automation Engineers, Plant Managers, Data Analysts, and engineers involved in Continuous Improvements and EH & S.
Part 1 - What is Manufacturing Intelligence?
Part 2- Data Sources And Transactions For Manufacturing Intelligence
Part 3- Manufacturing Intelligence - Data Storage
This topic is not limited to manufacturing and is useful for all facilities with a need to collect and use large amounts of automated data.
This blog series is about developing the infrastructure and selecting the tools that are needed to get business value from automated data collection at your plant or facility. Manufacturing Intelligence (MI) is the term used to describe the process and the components needed to obtain plant floor data and facility data from many data sources and make it available in a useful form, from the plant floor personnel to the enterprise. That “useful form” can take many shapes, such as being combined with other system data to bring more context to the data; being available in automated reports, visualizations, and data analytics tools; and being summarized and processed into dashboards. The data can also be used to feed machine learning algorithms to provide forecasts and predictive analytics to help optimize your process, provide notifications, and schedule equipment maintenance.
Below is an overview of MI. Depending on your role at your facility, you may be familiar with some part of the MI process, but not all. MI applies to all facilities with data, not just manufacturing plants. In the rest of this blog series, we will delve into the details of each part of MI, which are summarized below.
Data will be located in many types of devices in a facility or plant. This provides a challenge for obtaining the data since the correct drivers and software are needed to get the data and move it to a database. Here are some examples of devices and software that may have the data you intend to capture and log:
In addition to getting the data from the source, there needs to be a means to log the data to a database based on a trigger or to log data periodically. In some cases, previously logged data records may need to be automatically updated. For this you need a transaction software or data logging software. There are a wide range of choices available, including:
We will explore both the data sources and the essential data logging/transaction software in the next blog post in this series.
There are two forms of primary data storage that are used:
Many plant-wide data collection systems use both types of databases as they are not mutually exclusive. We will take a deeper dive into the database architecture in blog post #3 of this series and show the useful features of each database type for securely storing your plant data.
In many cases, there are requirements to interface with multiple databases or transform data within a database to make it more useful for users. Here database integration is used. Extraction, Transform, and Load routines (ETLs) are used to process data from a data source and send it to a data destination. These can range from relatively simple file read and import scripts to complex code using data integration tools that are designed to read data from multiple sources, aggregate the data, and fill a database designed for efficient reporting.
We will review some of the various scenarios in blog post #4 of this series to show how data integration is used to combine data from various sources and to improve report performance.
Data visualization is the part of the system that most people see, and it is the primary end result of all the back-end parts of the system.
There are many choices for data visualization, including:
In blog post # 5 of this series, we will explore the various choices in data visualizations, the pros and cons of each, and show you how to get the most out of your data resources using visualizations.
Steve has retired from Hallam-ICS, but his contributions to the company continue to be valued.
Read My Hallam Story
Hallam-ICS is an engineering and automation company that designs MEP systems for facilities and plants, engineers control and automation solutions, and ensures safety and regulatory compliance through arc flash studies, commissioning, and validation. Our offices are located in Massachusetts, Connecticut, New York, Vermont and North Carolina and our projects take us world-wide.