Dashboards deliver data-driven decisions

Nov. 1, 2014

A data island may be stumbled upon almost anywhere, and in any industry, but is found nowhere more frequently than in production plants and facilities.

By Llyod Colegrove

A data island may be stumbled upon almost anywhere, and in any industry, but is found nowhere more frequently than in production plants and facilities.

Many of the commercial goods important in our daily lives are made using highly variable ingredients and tightly controlled closed-loop processes. Results are dependent upon highly customized, or at the least, highly configured equipment. Each system, piece of equipment, instrument and even device tends to have its own database storage and I/O drivers.

And even if data nevertheless is easily extracted, it’s not information until placed into context and made available for analysis. Thus, without an aggregating system to analyze and deliver relevant information, the data’s value plummets almost immediately after its generation.

Dow Chemical of Midland, Michigan, one of the world’s largest chemical companies, with more than 5,000 plastic, chemical and agricultural products manufactured at 188 sites in 36 countries, is changing that, just as are global corporations and even small and mid-size enterprises everywhere.

At Dow, any given product is likely made at more than one plant location and done so using the systems, equipment and processes appropriate to that environment.

Visual in real time

Across the process industries, a long history of corporate mergers and acquisitions means facilities that are today part of the same company have different legacy IT and automation systems in place. Data collection and analysis is complex in itself, but comparing production facilities and product quality; identifying constraints; and developing best practices across multiple facilities can seem scarcely possible.

Yet millions of dollars aren’t available to revamp process-control systems or IT infrastructure. Upgrading or replacing systems is unrealistic because cost prohibitive, if nothing else, and often unnecessary.

In one such case at Dow, the challenge was addressed by enabling more general data access. But it was done without moving the data to some redundant, parallel data space. Databases remained in place, but access to the data where it already resided was eased.

Northwest Analytics NWA Focus EMI solution allows direct data access from existing data sources. By passing it through a global analytics layer, critical process information is "visual" in real time. Dow has the information needed to monitor, understand and control its processes.

The return on investment is substantial. The solution allows fast identification of supplier raw material changes that can negatively impact catalyst. It prevents fresh catalyst losses and has been associated with the longest run in Dow history between scheduled maintenance turnarounds. The solution is "agnostic," i.e., able to analyze data across operations, delivering a view of process health information in real time.

As my colleague, Mary-Beth Seasholtz, Ph.D, a senior scientist here at Dow has said, "Dashboards are shared across departments. When behaviors are contrary to agreed-upon metrics, the decision becomes data-driven."

Beyond intelligence

Besides "intelligence," the return on investment for this investment was immediate and even surprising. Quantifiable results based on direct access to data led to the avoidance of expensive mistakes and saving millions of dollars. The return on investment is estimated at $1,000,000 – $2,000,000 per plant per year.

Having intelligent "views" shared by professionals in different roles and global sites leads to agreement when it comes to metrics and variables. The "views" make information role-based and actionable, liberating it from constraints associated with spreadsheets stored on laptop hard drives.

There is no guessing about what the data reveals in terms of manufacturing processes or behaviors. Decisions can be made that are deliberative across company levels, eliminating finger-pointing and time-consuming remediation procedures. As was the intention, decisions are accelerated and data driven.

Dow gained more control over off-grade product and reduced raw-material waste levels. Better understanding of subtle variations in qualitative measures can significantly impact catalyst and filter activity, and lead to other instances of increased life cycles. Downtimes are reduced.

The system’s two most recently introduced models include NWA Focus EMI KnowledgeBase, which captures institutional knowledge — such as assignable cause/corrective actions — and best practices. The other is the NWA Focus EMI Collaboration Center, which speeds issue identification and resolution by enabling issue-related information, data and documents to be shared across the organization.

Llyod Colegrove is director of data services, Dow Chemical.

Northwest Analytics, headquartered in Portland, Oregon, delivers manufacturing and SPC solutions to improve plant processes in manufacturing enterprises and supply chains. Its emphasis is on integration, analysis and visibility and it has 3,000 customers worldwide.