Activplant Whitepapers (click to download)

ActivplantEnterprise ManufacturingIntelligence:

Introduction

Data collection, reporting and analysis are fundamental components of the manufacturing process. Together, they drive the decision making process, which ultimately impacts the bottom line. Still, remarkably, the ability for manufacturers to truly see what is happening in real-time across their entire operation has been, to date, limited, and, most likely, flawed.
Manual data collection, which has been the hallmark of many operations, is limited by what the operator chooses to see and record. The end result is decision making that is only as good as the information that was first recorded.

There is a solution, one that delivers powerful, useable actionable information capable of revolutionizing the way a manufacturer does business to deliver more complex products at a lower cost delivered exactly when customers need them. It’s called enterprise manufacturing intelligence (EMI), and it is delivered by Activplant.

The following paper outlines the issues that face today’s manufacturer, and details the critical components of a true enterprise manufacturing intelligence solution – a solution designed to give true operational visibility across multiple plants around the globe.

Enterprise Manufacturing Intelligence Defined
Issues for Today’s Manufacturer
More so than ever before, plant managers face tremendous stress to perform. The stress points include:

• Increased production – Manufacturers have to produce more products of higher quality without additional resources. It’s not enough to simply add more assets. It means the issue of poor asset utilization must be addressed.

• Lower costs – Cost reductions are a reality, as everyone in the value chain is demanding lower costs. The threat that the customer will find another supplier that will meet their needs is all too real.

• Complexity – Products being designed today, as well as the product mixes that have to be produced by the same equipment, are increasing in complexity. The result is plants that are far more stressed and must be far more flexible than ever before. As complexity increases, so do the challenges of finding ways to improve efficiency and reduce costs.

• Traceability – Customers are demanding increased accountability for parts traceability and birth history.

• Fewer staff – Fewer staff are available to monitor operations.

• Business systems lack accurate data – Planning and business systems do not have the real-time, accurate data they need to be effective. There is no connection between the plant floor and the business systems like ERP, CMMS and SCM. As a result, scheduling and planning are based on inaccurate information, which leads to cost increases and operational inefficiencies.

• High TCO – The costs are high to maintain multiple monitoring systems with multiple interfaces and databases that do not deliver the manufacturing intelligence managers need to be effective.

• Speed of program launches – There is a requirement for new programs to accelerate from zero to full speed production in as short a timeframe as possible.

• On-time delivery – Customers, especially lean manufacturers, are dependent upon on-time delivery to meet their production run requirements.
Activplant EMI White Paper 3
Problems with Current Approaches
In order to meet targets, plant management requires accurate, timely and complete manufacturing intelligence (MI). However, existing systems cannot deliver what the managers need because:

• Information is not timely, accurate or complete – Typically, data is not automatically collected, but is manually gathered – often on paper – and is, as a result, subject to operator error, bias and delay.

• Difficult to analyze – Existing systems are focused on little more than reporting and provide information that is difficult to drill down on and analyze. The information that is available is not presented as integrated, real-time and historical data with applied business rules in the form of key performance indicators that quickly and easily enable analysis.

• Cannot view from multiple perspectives – The information is difficult to view from multiple perspectives, such as by batch, lot number and shift. Information gathered lacks any context, which means it is up to the user to define how things relate to each other.

• Systems are inflexible – Change on the plant floor is constant and these systems require costly and time-consuming custom coding. As a result of both cost and time delays, these systems do not reflect plant floor reality. Furthermore, corporate management may impose standards across all plants to ensure there is a consistent presentation of information. Individual plants are then restricted to the corporate standard as opposed to having the ability to develop a standard that works best in their individual environments.

• High Total Cost of Ownership (TCO) – Existing systems carry a very high total cost of ownership, because implementation, modification and maintenance require either additional internal IT expertise or costly consulting. As a result, they are slow to implement and costly to maintain.

• Cannot scale – Today’s systems often fail to scale for use across the entire plant. Small systems that are restricted to a handful of machines do work – but just barely. They collapse when having to monitor dozens to hundreds of machines. They completely fail if they have to deal with multiple plants.

• Multiple systems – Typically, the plant has multiple, disparate and isolated systems that only deal with part of the problem – there is no mechanism to integrate the data gathered into an overall view. In fact, the only way to view results from across the plant is to make considerable investment in customization that enables an apples to apples comparison of the data. Multiple systems significantly increase the cost of ownership and present the problem of integrating many interfaces and databases into a single, unified view of the plant floor. This problem is even more significant when multiple plants are considered. It is not uncommon to have hundreds of non-integrated monitoring systems across a large corporation.

• Continuous Improvement and Six Sigma Projects lack key data – Managers responsible for Six Sigma and CI projects do not have the detailed manufacturing intelligence they require to undertake successful quality projects. MI is critical to be able to determine where the best opportunities for quality improvement exist. The right data generated by MI, more importantly, is key for the measurement and analysis processes in Six Sigma and CI projects. MI is also crucial to prioritize different projects for maximum return on investment, and to determine if a project was, in fact, successful.

Enterprise Manufacturing Intelligence –The Missing Link in Modern Manufacturing

Enterprise manufacturing intelligence is a critically important requirement in today’s highly competitive manufacturing environment. While manufacturing intelligence is not new, its focus has been on lines, departments and individual plants, leaving those companies with multiple plants searching for something that can deliver across multiple plants around the globe. Enterprise manufacturing intelligence is the solution.

It’s not enough just to have stores of data – data must be able to communicate and be actionable. In an EMI environment, data sources link with manufacturing and business logic and are the basis for actionable, strategic reports, analyses and key performance indicators (KPIs) such as OEE. Together, these functions arm each department in a manufacturing operation with the knowledge it needs to drive company goals.

With EMI, data is collected from across the entire plant floor or multiple plant floors, stored in a single database and is designed to provide one cohesive, integrated view of any manufacturing operation. Through strong reporting mechanisms, users can actually see what is happening on the floor, determine the trends and know how effectively the plant is running. What’s more, EMI closes the loop between the data generated on the plant floor and other systems used to drive the plant. Managers become empowered to act now having reports and KPIs on scrap, machine utilization, starved machines, quality and other key elements of the factory floor operations. This puts the power into the right hands and enables managers to affect change that results in increased productivity and quality, and reduced costs. In addition, business systems have the intelligence they need to efficiently schedule the plant, generate accurate activity-based costing, schedule effective predictive maintenance, and drive the supply chain.

While aggregated totals of production counts may satisfy corporate management, the ability to relentlessly drive out costs, improve quality and squeeze out more products using existing resources is predicated on the availability of intelligence from the factory floor. Anything less will give a misleading, inaccurate and incomplete view and will not bring the results demanded.

Every sector of the economy, particularly manufacturing, is facing extraordinary pressure to drive down costs. Increasingly, customers are demanding two to five per cent price cuts this year and will do so every year into the future. They are also looking for more complex products with increased quality. If you cannot satisfy them, they will take their business elsewhere. It is imperative that every factory, plant and shop find ways to produce more product, cut costs all without compromising quality.

The days of easily finding inefficiencies and cutting out costs, however, are over. Most companies have found the simple fixes. Now they need to dig deeper into manufacturing process to drive out costs. In order to do so, they need intelligence – fast, accurate, reliable, configurable, maintainable intelligence. They need to find the hidden constraints and the quality issues, the causes for which are buried in the noise, buffers that are too large, and in excess inventory levels that exist within any manufacturing operation. They also need to be able to have the 30,000-foot view of their plant floor – the kind of view that enables them to see how processes function together across the entire plant so they can affect the right change as appropriate. The only way to give them the information they need is via EMI.

The opportunities for finding hidden costs are not going to be discovered with stopwatches, clipboards, Excel® and an army of observers. Companies do not have the resources to devote to painful, time consuming and inaccurate manual data collection methods. Nor can they afford automated systems that require multiple products and databases and are, by their nature, incomplete.

Companies need to manage their assets more proactively. Too often managers find out about problems or deficiencies too late to do anything about them. Imagine the typical automotive plant where plant management gets the actual results of production the next morning. While the information represents interesting historical fact, it does not enable anyone to change the past. The only real value to this information is its ability to help a manager prepare excuses for what went wrong in the previous day’s production.

An effective EMI system will not only track production information in real-time, but will also alert staff to problems on the floor in time to do something about them. The system also gives a manager and his staff the opportunity to analyze the issue, drill down to the detail, identify the cause, determine what they need to do about it and take action when they still have time to fix the problem.

EMI BENEFITS

The benefits of a fully functional EMI system are substantial, and will impact the entire manufacturing organization. EMI will increase revenue through its ability to:

• Facilitate increased production – An EMI system enables a complete understanding of where and why downtime, poor machine utilization, scrap and other factors occur. That knowledge comes from the operational visibility that comes from an EMI solution. Having that knowledge along with the ability to react to these serious issues in time is critical to ensuring any manufacturing facility is capable of meeting the increasing demands of its customers.

• Lower costs – The sophisticated reporting and analysis capabilities derived from an EMI solution help target the source of problems quickly and efficiently and drive them out of an operation. An EMI solution also means manufacturers do not need to maintain multiple, costly monitoring systems with multiple interfaces and multiple databases. It’s a single solution that drives effective manufacturing intelligence for sound decision making.

• Handle complexity – The flexibility of an EMI solution gives manufacturers a competitive edge because it can be reconfigured quickly and easily using existing staff knowledge to accommodate a new product mix to deliver the operational visibility required for complex manufacturing.

• Improve traceability – The growing demand for product traceability requires the level of sophistication that can only be delivered with an EMI solution. Its combination of real-time and historical data delivers the birth history and traceability required.

• Reduce the dependence on manual monitoring –With an EMI solution, plant floor personal can attend to operational issues rather than doing time-consuming and cumbersome manual data collection. This saves time and reduces the human biases that creep into a manual reporting system. Automated data collection can also be more extensive and complete, than manual data collection.

• Track and store accurate, real-time data – Business and planning systems arenever starved for information from the plant floor with an EMI solution. That’s because the loop between the plant floor and the administrative offices is closed. An EMI solution feeds the business systems and enables solid, accurate decision making to occur. Managers can reevaluate production plans on a daily basis if necessary. They can also review standardized reports and develop standardized calculations across all plants, but can still offer a degree of autonomy to individual plants to help drive best practices.

• Increase the speed of program launches – Reducing the amount of time to commission new equipment during a program launch results in increased sales during the time period when a new product is in greatest demand. Critical to note is that with an EMI solution, manufacturers can drive lower costs far more quickly than with other systems.

• Ensure on-time delivery – An EMI solution allows a manufacturer to improve machine efficiency, remove system constraints, audit and verify inventory and expedite effectively. Because EMI allows plant floor personnel to identify problems in real-time, they can also be solved quickly to ensure the production and delivery schedules are maintained. EMI offers true knowledge, the most powerful component in the engine driving on-time delivery.

Ultimately these benefits have a critically positive impact on your manufacturing facilities. They enable companies to reach corporate goals such as higher throughput, reduced costs, increased quality and better on-time delivery.

Enterprise Manufacturing Intelligence – A System Defined

EMI, like an iceberg with 90 per cent of its volume below the water’s surface, consists of multiple layers hidden from view – layers not typically considered when an EMI solution is evaluated. But it’s these hidden layers of infrastructure that are the foundation for a successful EMI system.

In environments without EMI, management has to rely on a paper-based or static, custom data collection system to provide information about what’s happening on the plant floor. These outmoded systems deliver inaccurate, incomplete and delayed information restricted simply to monitoring and reporting. A system that only focuses on monitoring and reporting is like a car’s engine warning light – it only tells you that you have a problem. While finding a problem is a necessary step, managers are paid to deliver results.

A true EMI system enables you to find where the problem is, helps to identify its cause and alerts you in time to actually be able to manage it.

Each layer within an EMI system is composed of multiple elements that all have important contributions to make.

DATA FOUNDATION LAYER

The heart of any EMI solution is the data foundation layer, as it ultimately determines the value of the data, information and intelligence generated by the EMI system. It must collect, store and manage raw data rapidly, effectively, accurately and completely or the reports and analysis generated by the system will be flawed, incomplete and inaccurate. This layer is the most difficult and costly to deliver correctly. It must also be able to adapt to differing customer needs, such as a customer with thirty plants each producing different parts and differing business practices that need a single system.

Automatic Data Collection

The automatic data collection component of an EMI system is responsible for interfacing with the plant floor systems to automatically collect the needed data.

• The system must collect all the relevant data including:
• Production, scrap and reject counts
• Cycle times and counts
• Process parameters
• Downtime
• All fault/alarm/work stoppage/etc. incidents
• All process variables
• Monitor/track all buffers
• Monitor/track all events
• OEE and efficiency values
• Provide product and component genealogy
• Clearly identify production constraints
• Collect/track/analyze all quality parameters
• Support all relevant Six Sigma/ Lean/Continuous Improvement initiatives

Rather than relying on multiple disparate systems like historians and others to collect data, the automatic data collection system functions like all of those systems rolled into one. It enables you to add individual data points, assets/machines or even entire departments or lines full of machines without shutting down the collection engine and loosing any data. It provides alternatives for collecting data at the level of the PLC or in a central database, and enables you to add additional data collection points when the number of monitored assets increases.

Manual Data Collection

This component provides a mechanism for operators to augment automatically collected data with additional information such as reason codes for downtime occurrences and scrap.

Dynamic Data Modeling

Within an EMI solution, there is a single, unified dynamic data model for all installations with a single logical database. This approach provides a mechanism for collecting, storing and reporting against common data types. It also enables non-IT personnel to define the relationships between different data items in real-time such as by hour, shift, day, month, by any customizable time frame, model number, batch number, serial number, VIN number, and so on.

In order for the system to be truly successful, the data model must be dynamic, meaning it is possible to add all additional data collection points and types without requiring coding or database access and expertise to accomplish. It must also be able to be modified on the fly without stopping data collection.

The common data types collected include:
• Time stamp data, such as “Pressure at 12:08:45 was 175 psi”
• Time duration data, such as “Machine was starved from 1:08:39 to 1:20:00”
• Counters and timers, such as “The product count is 123”
• Context to tie all the data types together for viewing the data. Viewing a report by hour, batch number, and operator number are examples of context.
• Bill of Materials
• Product Specifications
• Routing tables

Dynamic data modeling is critical because it enables the EMI system to generate typical reports like:

• Corporate Management Reports
• Plant Efficiency
• Corporate Standardization with Plant Autonomy
• Plant KPIs in Real-Time
• Activity-Based Costing
• Quality Management Reports
• Quality Reports for Six Sigma and CI improvement projects
• Verify Six Sigma and CI Results
• Parts Traceability
• Scrap Reports
• Plant Management Reports
• Operational Visibility
• Real-Time KPI Reports
• Real-Time Production Counts
• Real-Time Inventory Reports
• Real-Time Status Reports
• Birth History and Traceability Reports
• Materials and Logistics Management Reports
• Real-Time Inventory Reports
• Scrap Reports
• Real Cost Reports
• Empower Customer Service
• Finance Management Reports
• Automatically Supply Business Systems
• Activity-Based Costing
• Production and Operations Management Reports
• Bottlenecks Improve Utilization
• Better Scheduling
• Scrap Reports
• Optimize Buffer Size
• Maintenance Management Reports
• Downtime Incidents in Real-Time
• Machine Cycles
• Predictive Maintenance
• Multi-Dimensional Analysis
• Feed CMMS and EAM Systems Real-Time Data

Manufacturing Metadata

The EMI system must be able to map customer terminology for data items such as scrap count to the terms used in the data modeling layer. It must also handle unique manufacturing characteristics like shift management, part numbers and asset hierarchies

Data Exchange

This particular system component is responsible for ensuring data can be exchanged between the EMI system, other plant floor systems and business systems in a format that is efficient for all. Given the multiplicity of systems that exist in most plants, it is essential that any EMI system be able to exchange data effortlessly between the plant floor and the business systems.

MANAGEMENT LAYER

A true EMI system has to deliver the security, data management, reliability, extensibility and standards support demanded by large-scale deployment both at a single plant and across multiple operations. Only a system designed from the ground up can provide the range of services necessary to provide the needed scale.

Configuration

Configuration is critical to the flexibility of the system, as it is responsible for the ease of implementing and maintaining the EMI system. Plant floors are, by their nature, dynamic and the system must be flexible enough to handle the changes that are an inevitable part of any EMI system. For this reason alone, it is imperative that the configuration tools support both drag and drop as well as the reuse of data objects and machine definitions by plant floor staff without requiring costly IT department personnel to support the system’s day-to-day use.

Extensibility

An EMI system must have the capability of being extended by third parties and customers to add further functionality and to leverage the data collected and stored by the system. This capability should include published APIs, a documented software development kit and training.

Data Management

The EMI system must allow customers to automatically archive, summarize and delete data based on customer-defined rules. In some industries, information has to be retained for 15 years or more so it is imperative that there is an automated, rules-based system for managing data. Without it, there is a risk that the data storage requirements will become overwhelming or that key data needed for customer or regulatory compliance will be lost.Standards-based Infrastructure Support

The system must integrate into existing corporate systems and leverage the investment already made. It must also support all the corporate standards such as OS, databases, information protocols and so on.

Reliability

The EMI system is a key to the success of the plant and manufacturing organization and must have the built in redundancy, fail-over and data integrity capabilities.

Security

An EMI should provide granular security down to individual data items and by user to allow any mix of security and user roles. The security system should also leverage any existing security systems to minimize the administration of security.

In addition, users should be assigned to user roles to reduce the administrative overhead of managing security. The system should support existing corporate security standards and security information stores and allow the importation of existing security information. Organizations should also have the flexibility to choose the level of security they want to use and have controls available to restrict users ability to modify data.

DEPLOYMENT LAYER

The deployment layer provides the services necessary to distribute the reports, analysis, alerts and KPIs across a plant or multiple plants. It defines the scalability of the solution and the way in which it easily it can be deployed across an entire manufacturing organization.

Multi-plant deployment

This component enables customers to administer and deploy EMI capabilities across multiple plants while maintaining the autonomy of each individual plant to develop its own reports, analysis and KPIs.

Multi-plant deployment allows the corporation to establish and reuse standard templates for calculations, data collection point definitions, graphics, reports, KPIs and alerts across multiple plants. It simplifies and standardizes the way in which the solution is implemented at a plant level within the guidelines and framework established at the corporate level. This approach enables each plant to leverage the standard elements and templates that were developed at the corporate level and allows for rapid implementation of corporate-wide changes to the system.

Critical to the success of multi-plant deployment of an EMI solution is an environment that both enforces conformance on key issues, calculations and KPIs, and provides all the flexibility plants need to implement individual solutions required specifically in their environments. This approach greatly accelerates implementation of an EMI solution and helps guarantee report standardization across plants.

Distributed Architecture

A distributed server architecture ensures the EMI system is robust and able to provide the mission-critical information needed to run the plant. It also helps to ensure the system can scale from a small plant with dozens of assets to the largest and most complex factory with thousands of assets, hundreds of users and millions of data points.
Distributed architecture also means customers can install the solution on a single server or across multiple servers with fail-over capabilities. In addition, the architecture can handle multiple plants across an entire manufacturing organization.

Web-Based Server

Effective EMI solutions must be web-based to minimize deployment and implementation costs.

Zero footprint Web Client

The interface point between the user and EMI system must have a zero footprint to minimize the cost of deployment, implementation and maintenance. Without a zero footprint client, deployment issues may severely limit how widely the EMI system can be distributed.

DATA LAYER

The data layer, in combination with user-defined static and ad hoc reporting tools, provides users with real-time and historical data reports. Users must be able to view reports from a single asset, line, department, plant or series of plants.

Real Time Data

The EMI system must enable users to view reports run against real-time data.

Historical Data

The system must collect and report against historical data with any degree of summarization and time periods of up to 15 years.

User Configurable Static and Ad Hoc Reporting Tools

The EMI system must provide user configurable tools to define, modify and view new as well as existing static and ad hoc reports using a web browser. Users must be able to save and view these reports at their discretion. The EMI system must enable users, with a single click of a mouse, to export data to MS Excel® for further analysis and reporting.

INFORMATION LAYER

The information layer enables the user to see multiple data types in a single report. This is important for trending and other reports where a combination of data needs to be consolidated to provide meaning.

Data Integration

The EMI system must integrate data from multiple sources – both real-time and historical – in a single view with a single, integrated context. For instance, the system should be able to show both real-time and historical production data for a particular lot number or part type.

INTELLIGENCE LAYER

This layer provides the analytical capabilities to explore and find root causes, as well as intelligent discovery through alerts and analytical applications, such as build-to-order systems, error proofing and activity-based costing. It works with the analytical tools to enable users to drill down in detail from a high-level summarized view to explore the detailed data to find the cause of the problems.

Business/Manufacturing Rules

Users must be able to create powerful, standardized business rules and calculations to ensure a standard method of information presentation. The system must also enable users to create their own rules without having to resort to a programming language.

Visual Key Performance Indicators

KPIs are an aggregate of multiple metrics combined with a visual layout that enables management to understand, at a glance, where they need to focus their attention. KPIs also provide a powerful insight into the key aspects of the plant. Key Performance Indicators also help accelerate management’s understanding of where problems exist.

Automated Discovery – Alerts

Alerts enable users to create highly configurable events and customizable alarms for either themselves or the entire plant. Notification of issues as they arise frees users from having to constantly monitor their lines.
The system’s architecture must enable users to develop both personal and global alarms and have the capability of handling the complexities of the plant floor. The ability to generate an alarm or undertake an activity comes from a sophisticated set of rules based on simple and complex triggers, times, calculations, measurements and counts.

User-Configured Analysis Tools

Users can easily view reports from multiple dimensions, such as by batch, shift, VIN number, line and more, and then drill down to find the root causes of the problem. Since each user has a different requirement for information, Activplant enables users to create the reports and views on information they require to do their job. Rather than having to rely on IT personnel to create this information with a resulting delay and cost, users are empowered to create their own reports.

Analytical Applications

Analytic applications are built around standardized reports and best practices and are used to drive a higher order of understanding related to processes and equipment. This requirement necessitates a highly flexible and configurable data visualization environment that facilitates simple how-to and what-if scenarios. Analytic applications can also play a role in execution. Error-proofing systems, for example, require the manufacturing process to make production line decisions based on collected data. Analytic systems can help achieve this high-level requirement.

ENTERPRISE ADMINISTRATION LAYER

This layer provides the administration tools that make it easy to manage the multiple elements within the EMI solution.

Summary

Enterprise manufacturing intelligence is a critically important requirement in today’s highly competitive manufacturing environment. In an EMI environment, data sources link with manufacturing and business logic and are the basis for actionable strategic reports, analyses and KPIs such as OEE. Together, these functions arm each department in a manufacturing operation with the knowledge it needs to drive company goals.

In environments without EMI, management has to rely on a paper-based or static, custom data collection system to provide information about what’s happening on the plant floor. These outmoded systems deliver inaccurate, incomplete and delayed information restricted simply to monitoring and reports. A true EMI system enables manufacturers to find where the problem is, helps them to identify its cause and alerts them in time to actually be able to manage the problem.

Activplant is the only EMI solution with the complete mix of functional layers to deliver the information that both raises questions and helps answer them.

In short, Activplant generates results.