Physical assets used in the process industries can range in scope and size from a single instrument or controller to a full-plant or multi-plant complex. These assets represent a significant investment and it is generally accepted that their respective lifecycles must be managed to achieve optimal return. As discussed in detail in previous ARC reports, overall asset lifecycle management (ALM) involves a set of interconnected, iterative processes:
• In the design and build phase of a project, programs are managed for project performance (PPM).
• During the much longer operate and maintain (O&M) phase, the process focuses on asset performance management (APM).
• Asset and project portfolio management (APPM) aligns the overall portfolio of investments in assets with the company’s strategic objectives.
Model for ALM/ALIM Processes
Because of their interconnection, these processes require a continuous exchange of information. This is achieved through asset lifecycle information management (ALIM). ALIM incorporates the collection, management and distribution of information on design and construction, operation and maintenance of the asset portfolio and makes the data available to PPM, APM, and APPM.
The overall purpose of these processes is to maximize the net total value of ownership over the asset's lifecycle. Yet the multi-parameter optimization for this process typically involves tradeoffs because of parameter interdependency. For example, if a plant is built faster and is in operation earlier, the project costs may be higher, but this is more than compensated by the additional value of products sold. And while maintenance costs can be reduced, this also likely results in reduced productions and can possibly shorten the lifetime of the asset. Organizations have opportunities to exploit this potential for global optimization. To reach the optimum, timely, in-context, high-quality information is required and must be transparent from the controller or sensor up to the highest level of optimization.
ALIM Data Quality
According to Milo Scheeren, Chief Product Officer of BlueCielo, there is an ALIM data quality risk. Among other issues, low data quality can result in non-compliance with regulations, or create a safety risk, as staff may make wrong decisions based on incomplete or incorrect data.
Full remediation can be expensive, as inspecting data for quality is tedious and costly. Scheeren recommends that focusing on the most important data might be more affordable and effective. He recommends that owner-operators structure data and compare them with reference data libraries from ISO standards. Dashboards representing data quality can help owner-operators assess urgencies and gaps and used as a basis to create a plan. Data quality management can provide benefits in the range of 2 to 5 percent of the maintenance budget, in addition to longer term savings, says Scheeren.
Peter Spelberg-Jahn, product manager for Siemens COMOS agrees with the importance of ALIM data quality. He notes that only important, often-used data needs to be structured. Rarely used, auxiliary data can remain unstructured. The challenge according to Spelberg-Jahn is to know what is important and what isn't. Experience shows that this is highly dependent on the company and culture and cannot be derived in a straightforward manner from equipment categories or other PLM criteria. This means owner-operators need to asses which equipment or data have most impact on their performance based on their operating and maintenance strategies.
And what about the consequences of wrong data? Spelberg-Jahn explains that an object-oriented centralized data hub will show inconsistencies and forces the user to decide which data is wrong and which is right. Good asset information management throughout the lifecycle can save up to 50 percent of engineering and maintenance time, according to Spelberg-Jahn.
Predicting Asset Performance Degradation
Mike Brooks of Mtell explains how a new analytics solution can predict performance degradation of equipment weeks in advance using weak signals that people would not have picked up on. In his experience, by the time problems are noticed, some damage has often already occurred. This compresses the time available to apply maintenance, leaving little option for scheduling both maintenance and related operational activities. The solution can also prescribe how to maintain equipment or adapt processes to avoid deterioration. He also reports cases in which users did not trust the first detection of an anomaly, ultimately resulting in equipment damage. Brooks uses the term “process abuse” to indicate how the process can inadvertently be pushed for performance reasons without understanding that this can shorten the asset longevity.
Processes need to be changed to avoid recurring damage, he insists. He reports a case in which the software helped reduce downtime from 72 to 94 percent. A transportation company applied the solution and gained $10 million in just three months. By extending the scope, the benefits amounted to $200 million in two years.
Evolving Automation Pyramid
Yosuke Ishii of Yokogawa explains how the automation pyramid is changing with new production methodologies such as modularization and Internet of Things (IoT) applications. An increase in both high- and low-level data creates opportunities for owner-operators.
Ishii mentioned that Yokogawa has identified nine use cases, most related to mobile maintenance solutions. Customers have shown interest in combining augmented reality and mobile solutions to access equipment data, ERP data and maintenance management information at the point where the equipment is located, thus helping to connect organizational sites.
ARC Recommendations
As managing assets in industrial facilities becomes more complex, both from technical and organizational perspectives, owner-operators, engineering procurement and construction company’s (EPCs), and other stakeholders should carefully design their collaborative processes. All parties need to define their responsibilities related to maintaining and transferring shared asset information and agree upon common quality standards. A strategy for the handover and exchange of information during the lifecycle must be defined using the available ISO standards (15926, 18101, 14224).
High-level processes must be complemented with modern applications, such as predictive and prescriptive analytics of both small (individual equipment-oriented) and big (equipment park-oriented) industrial data, mobile applications, new integration schemes and technologies and be fueled using more abundant sensor information in close to real time.
These measures should help optimize the total value of ownership or profitability for all stakeholders. They should be practical and effective and not be hindered by rigid principles or beliefs. For example, Industrial IoT (IIoT) applications can complement existing automation networks and existing IT applications in plant or enterprise environments.
