Big Data can help synthesize information and provide users with the exact data they need to make informed, intelligent decisions. Developing a process that benefits everyone in the plant from the top on down is crucial.
There is a tremendous amount of information out there; dissecting the information can be overwhelming. Especially when the information comes from different sources and may not be accessible to everyone who needs it. Big Data can help synthesize this information and provide users with the exact data they need to make informed, intelligent decisions. Knowing how to capture the Big Data and make it useful is the key.
In an enterprise manufacturing system, it is important to be able to see the whole picture when evaluating a problem. An operator on the plant floor can use supply chain information to understand upcoming production schedules. Executives can use the production capacity of their various plants to shift production overloads from one location to another. Just look at trying to estimate a plant's production. The company would need information about production-line capacity, warehouse capacity, personnel utilization, and sales forecasts to name a few. All of this data comes from different sources. Individuals throughout the company work many hours building reports in different formats.
Manufacturing business intelligence software is designed to help simplify this process. The business intelligence systems available today provide the ability to bring real-time and business data together into a centralized location and allow users to align data collection with goals and objectives. Reports and information generated from these systems are standardized, repeatable and can be available enterprisewide. These business software systems provide data aggregation, actionable alerts, and predictive analysis. They allow users to optimize the information and allow users to create business decisions quickly and intelligently.
The business intelligence software systems include both data collection and visualization, but the key item that brings everything together is the abstraction layer. This helps normalize manufacturing process data and make the information relevant to the operational teams. This software is normally a configuration utility or modeling tool which allows you to model and contextualize data pulled from the many disparate systems throughout your manufacturing process. ISA-95 modeling standards were followed in the development of these tools. It is best practice to utilize these standards when developing a system.
For example, when collecting data from a production process such as filling, there may be several types of fillers enterprisewide, and the users may want to calculate overall equipment effectiveness (OEE) for each. Even though the various fillers may have different inputs and outputs they have the same basic properties. When a filler is defined, the user will want to know the capacity, number of items filled, and downtime. The user will also want to calculate planned production time, which would require production schedules and personnel availability. Lab or quality data could be used to help calculate rejected or scrap material.
The image diagrams use of the abstraction layer to consolidate data from different sources. Instead of having several bits of data in different locations, all the information about the equipment is combined within a filler object. Data is aggregated into useful information that will now be accessible enterprisewide.
A major complication with making Big Data useful is the expertise that is needed from so many different resources throughout your manufacturing process. Organizing the data within these model-driven software systems provides a collaborative and secure environment were data is accessible to everyone who needs it. Users will have the ability to access the data they need through pre-developed dashboards and reports or via ad-hoc capabilities using Microsoft Excel and other client tools for creating trends and charts.
Successful implementation requires a team effort and all users to buy into the system. Find champions in various areas of the process to define business requirements and to model the data so it makes sense to everyone. It is best to start small and introduce new data or new functionality slowly. This will insure the system will be useful for everyone—especially when the company starts considering multi-site implementations.
This post was written by Patty Feehan. Patty is a senior developer at Maverick Technologies, a leading automation solutions provider offering industrial automation, strategic manufacturing, and enterprise integration services for the process industries. Maverick delivers expertise and consulting in a wide variety of areas including industrial automation controls, distributed control systems, manufacturing execution systems, operational strategy, business process optimization, and more.