The Answer is Hidden in Plain Sight
“So come close…because the closer you think you are, the less you’ll actually see…the easier it will be to fool you.”
The movie, “Now You See Me,” provides excellent insight into how businesses today focus on data and systems. Complex interwoven processes and systems, audit requirements and Sarbanes-Oxley compliance have businesses hyper-focused on intricate details. Heightened regulations increasingly emphasize capturing more detail at the lowest possible level. All details need to be captured, tracked, and cross-referenced. The result across the oil industry is that managers, operators, business analysts, and IT systems analysts chase these details, meticulously entering them into the respective systems. They are so focused on the details that they begin to lose sight of the larger trend, purpose, and implication in what they are doing. People are trained to work the screens and feed the systems. However, they can lose the ability to harness the data and glean insight when they are overly focused on minutia.
Traditional Asset Management Requires Constant Care
As one of the key links in the oil industry value chain, refineries consist of multifaceted interworking systems that function together to take in crude slates and produce a variety of refined products. Keeping a refinery running requires constant maintenance and supervision. Personnel are highly trained on all types of processes and equipment in a refinery to ensure they are functioning safely and efficiently in order to avoid downtime.
Refiners implement Enterprise Asset Management (EAM) systems in order to track and manage the complex refinery systems. The EAM system demands precise data and discipline in order to model every building, unit, piece of machinery, and critical part. The master reference data setup tracks the connectedness of each of the pieces of equipment as they join together to serve a purpose. The physical attributes of the equipment are tracked and managed (e.g., supplier, manufacturer, installation date). Updates and repairs are documented.
On a daily transactional level, plant operators and supervisors utilize the EAM to enter notifications to track when a piece of the equipment needs to be examined. The notification has a life cycle and process around prioritization. It allows supervisors to weigh in, offer their expertise, and progress the work to an eventual work order. The repair work is resolved internally, if skilled resources exist on staff. Otherwise, vendor management modules allow for the plant to seek bids, compare, and select outside skilled vendors to perform the necessary work. Planned and actual costs, length of project time, materials involved, and final completion are among the myriad of details tracked.
All along the way, each step requires meticulous feeding of the EAM system. Plant operators, managers, and supervisors must work according to complex processes in order to keep the plant operational.
Is Anyone Really Gauging the Health of the Plant?
With such a diligent focus on the systems, it is no wonder that businesses regularly find themselves struggling to track and manage the costs associated with assets. Management teams are often surprised by the immediate and critical maintenance required. The high costs create budget overruns from month to month. Vendor use creeps upward as supervisors rely more on outside personnel to do specialized work. Parts begin to accumulate in the warehouse as extras are ordered based on recent events and what individual supervisors feel is needed, rather than on pure empirical analysis.
Many organizations use simple time-in-service recommendations for asset maintenance lifecycles. While at times effective, it is rarely the most efficient means.
How to Pull Away from the Details to See the Larger Picture
Companies cannot ignore the importance of details for the EAM systems. The systems still require thorough detail in order to accurately track and manage the lifecycle of work orders. Detailed costs and hours still need to be tracked against the transactions. However, sometimes companies need to step back and adjust perspective; they need to make the data work for them. To accomplish that, the answer lies with business intelligence and a focus on the analytics that supply the larger trends and analyses on top of the EAM data.
Put Reference Data to Work for Your Needs
Location, location, location. It is important to get the best use out of implemented software. Inherent in any system is the reference data that maps the real world into a logical one. The plotting of data for spatial understanding is critical to insight. That data can be put to good use to create a locational analysis to help discover and visualize processes. An example would be utilizing visualization tools to plot data on top of asset schematics. Doing so promotes a deeper understanding of the asset.
For example, you can learn:
• Which types of equipment are found throughout various areas of the plant
• Average time-in-service of equipment
• Previous repair dates for equipment
• Which manufacturers are used through which sections of the asset
Apply the Right Algorithm to the Locational Analytics for Deeper Insight
Go beyond the basic reference data mapping and equipment setup. Once a company gains a basic insight into equipment, notifications, and detail of subsequent work orders, they must layer in the right algorithm. They should choose one appropriate to the asset’s needs. A common choice is the Weibull Analysis. Named for Waloddi Weibull, who published it in 1951, the analysis focuses on the versatility of distribution in order to determine reliability . It has become a recognized industrial standard used to determine the likelihood of failure given a small representative sampling of data.
Gone are the days of plotting points individually by hand. Now, through the use of advanced visualization tools, the Weibull Analysis can be incorporated and interfaced with the EAM tool.
The Weibull analysis provides insight into:
- Likelihood of failure
- Quickness of failure
- Clusters of similar equipment failures
- Failure modes
– Early mortality
– Random failure
– Early wear out
– Old age wear out
Engineers can then incorporate this analysis into their plans for the maintenance of critical parts and the concentration the skilled resources for an asset. Larger patterns and trends will emerge and additional questions might arise:
- If some failures are noticed for a type of significant part, how many additional failures might be expected over the next six months?
- How soon until the next failure might be expected?
- Is the rate of failure going to decrease (burn-in) or increase (early wear-out)?
- How many spare parts should be kept on hand?
- If most of the assemblies containing the part at risk are scheduled for routine maintenance in six to eight months, can the part be replaced at this time, or does a special shutdown need to happen?
Taking the analysis to the next level applies historical costs — both planned routine maintenance and unplanned critical costs. These curves, plotted against the Weibull analysis, provide further empirical insight on the frequency of maintenance versus its cost. The ultimate goal is locating the intersection of maintenance at the optimal time and at minimal cost for parts and equipment deemed most critical.
Turning Analysis into Actionable Insight
The possibilities become almost limitless for ways to slice the data that focus on the reliability of an asset at minimal cost. For example, from this base level, management can focus on:
• Vendor management and spend
• Planned versus actual costs for accuracy in budget forecasting
• Real-time analyses of critical equipment
It becomes a task of incorporating knowledge and analysis into regular management decisions and processes for continual improvement. This more advanced analysis is accomplished by utilizing the master data and notifications that are gathered as part of the standard asset maintenance process. Pulling value out of existing data and processes is key to continuously building enterprise value.