Advanced Analysis Using Unit Data and Event Templates
I believe you have to be willing to be misunderstood if you're going to innovate.
In this chapter, we follow the refinery digital transformation team as they find more innovative ways to use the enterprise industrial data infrastructure (EIDI) information for continuous operational improvement. This chapter ties together many of the concepts described in earlier chapters, such as event frames and offline analysis, as the team attempts to understand and resolve a gap between scheduled production targets and actual production output.
They design and prototype a simple template, known as the digital plant template, as a basis for tracking the mysterious variance. The template components analyzed are simply production rate and consumable variables through various stages of the refining process. The team generates EIDI event frames to track for these operational stages, which when reviewed, determine exactly when and where these hidden production and energy losses occur. After pilot testing, they now have a reusable template that they can deploy to the entire refinery on a unit-by-unit basis. This chapter outlines four key steps the team deploys to solve this problem, significantly accelerating EIDI time to value.
This chapter describes a more advanced solution to analyze trouble time, idle time, and downtime, through the use of event frames and the digital plant template. This chapter will likely appeal to readers who are experienced, proficient EIDI users. If you are not quite in that category yet, we hope you will take away some ideas on how to better analyze a refinery or plant process through the tools mentioned here.
This chapter explains and sequences through most of the techniques required for this analysis. At the end of the chapter, there is a link to a manual that describes these techniques in more detail.
Innovative Use of EIDI Capabilities
Peter Argus' digital transformation team met to review the enterprise industrial data infrastructure (EIDI) deployment progress. Other than some small hiccups usually associated with installing a system of this scope, the initial implementation was going well.
After the initial use assessment part of the discussion concluded, Tom Jordan, the refinery plant manager, commented that the Proclndustries South Texas refinery financial systems do not have sufficiently granular information about key operations production and consumables data. Much of this data is stored in the EIDI. Events that occur during the production process may shed light on causes of production losses. These events could be equipment startups or shutdowns, production set-up times, process idle times or downtimes, and unscheduled equipment shutdowns. Tom commented, "If we had a way to capture these events with relevant specifics, we could improve the overall flow of material in our refinery" (Bascur and Kennedy 2001; Bascur 2019; Bascur and Soudek 2019).
Tom told the digital transformation team that he wanted to capture these events in an effort to quantify the operating expenditures (ОРЕХ). "This will eliminate what we have been trying to do for years in spreadsheets. If the EIDI can do it automatically, the need for individual spreadsheets will disappear and the information will be available to everyone," he stated. He then presented a chart showing typical Proclndustries operational costs (Figure 4.1).
Tom explained that fuel is the largest operational cost followed by maintenance costs. The transition to a condition-based maintenance schedule is
Proclndustries refinery operational costs by percentage (HC loss = hydrocarbons lost during the refining process).
Four steps toward maximizing yields and reducing operating costs. (Courtesy of O.A. Bascur, OSIsoft LLC.)
helping reduce costs. EIDI-related data could potentially provide insights into other hidden energy costs.
Monica Armstrong (planning and economics coordinator), who helps define the daily production schedule, explained the current process to define refiner)^ targets and key performance indicators (KPIs). "The team needs to show management how they can use the EIDI system to apply new methods of analysis that can identify process improvement opportunities— specifically, tools that analyze data during specific operating modes of plant equipment and calculate comparisons between planned targets and actual production yields."
Peter and Monica realized that they need to have operational events classified in real time and quickly calculate production losses. By possessing this real-time data and the production loss times, they can improve on the parameters used as input to the refinery linear programming (LP), used for scheduling and optimization.
The transformation team constructs a four-step process to show management the flow of how the EIDI data is used to analyze and solve problems, as shown in Figure 4.2.