The Right Side of the Pyramid

The right side of the pyramid in Figure 6.1 represents business applications, big data, and operational intelligence support that can be used to increase the scope of the critical data infrastructure software. Much of this represents the IT side of the business. "Planning and scheduling activities are usually set from integrating the plant industrial data infrastructure with our enterprise business systems, which have production plans and utility costs," stated Chuck.

Layer 1: Condition-Based Monitoring

Layer 1 on the right side of the pyramid depicts the support center, which provides actionable information by all functions to improve the current state of the enterprise. For stable process plants, process equipment must maintain its highest availability in harsh industrial processing environments. Until reliable technologies became available, fully functional condition-based monitoring was not deployed.

Today, reliable process and equipment data streams enable maintenance to assign the necessary context to families of equipment. Thus, the required rules can be set and process analytics can be performed to classify the data to generate alerts and notifications, indicating a possible unscheduled shutdown looming ahead.

Underperforming regulatory controls and multivariable control loops cause process variability that adversely affects profitability. Having the data available in an easily accessible format for advanced analysis simplifies the continuous improvement required to support process controls at all levels for all plant process units.

Layer 2: Event Management

Layer 2 on the right side of the pyramid provides the support to validate and classify the data to develop information needed by continuous-improvement teams. These teams identify opportunities to ensure that all plant equipment is running or available to be run, with communication smoothly flowing among workers. Using online process analytics, the team can derive operational-mode time intervals in a plant and evaluate production and operational costs on a shift-by-shift basis. The transformation of data into actionable information using operational events to obtain production, energy, water, reagents, and other variables at the adequate degree of detail is essential. Transactional systems can be integrated with operations and production data to proactively improve overall production performance of the process plants (Bascur and Aroqui 2014; Bascur and Halhead 2013).

Layer 3: Performance Monitoring

After condition-based monitoring and assessments, performance monitoring (layer 3 on the right side of the pyramid) provides the final results. These results are transmitted to enterprise resource planning (ERP) systems to report all yields, product quality, operating costs, equipment availability, and inventories. Performance monitoring uses the advanced analytics of event data management to disseminate the data into actionable information. At this level, soft sensors can be implemented using predictive analytics, which are explained later in the chapter (Steyn et al. 2018). The physical sensors are backed up by the soft sensors, increasing process control robustness and stability.

Proper data classification and aggregation at the desired level of detail (to calculate all key operational metrics) enables faster communication and collaboration within the functional teams in the refinery. For example, it is always a delicate balance between the operations unit excessively pushing the equipment, and the maintenance team excessively maintaining the equipment. As described in Chapter 4, the data classification capabilities permit online mass balances to predict yield and run sensitivity analysis, thus increasing the value of the entire system (Bascur and Soudek 2019).

Layer 4: Planning and Execution

The analysis of information for overall production profitability, based on supply and demand, are provided by the planning and execution teams (layer 4 on the right side of the pyramid). Standardizing the information enables fine-tuning it for optimization tools. Planning and execution provides targets and schedules for the optimal refinery operations. Having the data infrastructure capabilities in layer 2 enables company personnel to quickly reduce operating costs. It also has enabled cloud-based technology where a production or manufacturing company can directly provide third parties (suppliers, service providers, domain experts, etc.) with highly contextualized data through a secure real-time cloud connection. An additional layer of analysis and support through equipment maintenance, catalyst provider support, water management, and outsourced external support has become a reality and will be further discussed in Chapter 9 (Bascur et al. 2016; Cope and Chugh 2019).

Accessing or sharing data with external services augments the acquisition of knowledge and extended support of remote plant operations. The collection and contextualization of real-time data is essential in providing input to more sophisticated offline predictive analytics and data analysis tools. Using the asset and time context in the EIDI allows for production event generation and provides aggregated data at the desired level of detail.

 
Source
< Prev   CONTENTS   Source   Next >