Structured method to collecting equipment data drives consistency in failure reporting, robust analyses

Methodology includes defining boundaries, constructing taxonomy for equipment, helps with proactive identification of reliability issues, corrective actions

By Charles Yang and Ian McWilliam, National Oilwell Varco

Equipment data is instrumental to understanding equipment reliability and performance, which necessitates developing robust solutions through root cause analyses (RCA) and effective corrective actions. The direct results are operation optimization and equipment reliability improvement. Drilling equipment reliability is the ability of the drilling equipment to perform its function within specified design parameters. By definition, equipment reliability directly affects operational efficiency and effectiveness, which affects the overall cost of ownership of a piece of equipment and the cost of drilling operations.

Benefits of Quality Data Collection

The drilling industry has seen great results from both drilling contractors and equipment suppliers utilizing their domain knowledge to improve operational safety, profitability and equipment design and performance. National Oilwell Varco (NOV) believes that quality data can impact the bottom line further, which is particularly relevant in the current cost-conscious drilling environment.

Quality data includes equipment maintenance, performance, failure and downtime data that can provide insight on how the equipment works and how the equipment interfaces with the rest of the system. Data can be aggregated and analyzed, with the results of the analysis facilitating proactive identification of potential issues and failure trends. The product of analyses depends on the quality of the data collected, with structured and consistent data being the foundation of insightful failure reporting and analysis. Along with robust RCA, failure cause is understood. Positive corrective action can be implemented to address and mitigate operational issues and improve equipment performance and design. Further, results of the analyses may help to form an effective preventive maintenance strategy.

Better data also helps to improve equipment safety and reliability, while increasing productivity and operational efficiency. Large volumes of data will help to accurately identify root causes of failures, issues and effects in near real time.

Challenges of Existing Data

Within existing equipment performance data being collected, some data entries are captured in unstructured text descriptions, comments and notes. This includes free text information that focuses on what was done to address an issue, instead of listing the failure or maintenance information in structured attributes. This approach does not methodologically document the components that are affected, why the failures or issues happened, or the effect of the event on the system.

As a result, it is challenging to perform comprehensive and systematic data analysis through this type of data. It is also difficult to directly extract meaningful and in-depth insight into equipment performance with such data.

Further, a substantial amount of work is involved in interpreting free text and comments when trying to gain insight from such records. The interpretation typically requires reading through the issues in text format, which can be time-consuming, and attempting to derive meaning from them can introduce errors. Due to the inconsistency involved, it is also challenging to compare component data across multiple equipment models.

Further, the current methodology lacks the capacity to automatically flag issues in the tracking system. There is no easy way of performing insightful, in-depth trending analyses on equipment failures.

To address the subjective nature of data quality and the limits of current data management systems, NOV determined that a consistent, more robust and intuitive methodology was needed. In such a methodology, it would be easier to follow a structured approach and mechanism for failure reporting, flag reoccurring issues in the tracking system, identify insights on failure trends, and quickly analyze patterns in impacted components or failure characters to optimize maintenance and reliability practices.

Improving Data Collection

Figure 1: By defining boundaries, a distinct description and agreement of which sub-systems are included in the scope of a particular system is established. NOV’s boundary-definition methodology dictates that an equipment family generally shares the same boundary definition. For example, all NOV top drives use the same boundary definition.
Figure 1: By defining boundaries, a distinct description and agreement of which sub-systems are included in the scope of a particular system is established. NOV’s boundary-definition methodology dictates that an equipment family generally shares the same boundary definition. For example, all NOV top drives use the same boundary definition.

As NOV examined maintenance and downtime records, the company identified key opportunities for facilitating information exchange and more productive information interpretation. Internal efforts primarily focused on improving the collection and exchange of reliability and maintainability (RM) data, with NOV determining what classification, or taxonomy, was required as a foundation. It was determined that there was no wrong or right way of establishing equipment boundaries and taxonomies but that it was imperative to keep a consistent methodology and approach.

NOV’s approach and methodology follows the ISO 14224:2016 international standard. The ISO 14224 standard, “Petroleum, Petrochemical and Natural Gas Industries – Collection and Exchange of Reliability and Maintenance Data for Equipment,” advises data standardization and exchange of information between relevant parties, e.g. plants, owners, manufacturers and contractors.


A boundary outlines the sub-assemblies in the system. The goal of the boundary definition is to establish a distinct description and agreement of which sub-systems are to be included and not included in the scope of the concerned system. This is the first step toward defining the taxonomy of the equipment. For each equipment class, a boundary will be defined indicating what RM data are to be collected.

Key aspects of NOV’s boundary-definition methodology include:

• An equipment family generally shares the same boundary definition. For example, all NOV top drives use the same boundary, regardless of being a hydraulic-driven model or electric-driven model, of different sizes, etc.

• Similarly, front-end control systems, such as Amphion and Cyberbase, and power systems, including electric and pneumatic, are typically outside of the equipment boundary. For example, the front-end control systems for both mud pumps and hydrarackers are outside of the equipment boundaries.

• Common boundary definitions are centered on functionality, not drawings. If an item controls a single function, it resides with that function; if an item controls multiple functions, it resides within tool controls.

• Naming conventions are kept consistent in all equipment. This avoids confusion by the end users of the database in the data-collection phase.


Figure 2 shows a portion of NOV’s top drive taxonomy. Field use was an important driver for this taxonomy development.
Figure 2 shows a portion of NOV’s top drive taxonomy. Field use was an important driver for this taxonomy development.

According to ISO 14224:2016, taxonomy is a methodical organization of objects into hierarchical groups based on shared attributes and characters. Locating information is more intuitive as the logical progression is tuned from overall focus to a more explicit focus. Subjects and components are grouped and defined in the context of their exclusive “parent-child” associations. Although there is no absolute right or wrong way to construct the taxonomy for equipment, it is imperative that the methodology be consistent. One of the results of having a consistent methodology is that it is intuitive to compare and analyze component data across equipment models and equipment families.

Field use is an important driver and consideration for NOV taxonomy development as part of a broader data-collection and analysis framework. The company’s taxonomies support the following goals and objectives:

• Standardizing language across organization. An example is that all drive systems should be called “drive motor assembly” across equipment families;

• Standardizing data reporting and collection;

• Easy to search; and

• Failure and trend analysis.

Equipment taxonomies are created with naming consistency and subsystem group consistency. This is essential when such taxonomies are utilized by an internal organization from engineering to field services. The taxonomy system is structured to be simple, straightforward, intuitive and easy to use. Processes are established to ensure all NOV rig equipment follows the same methodology when constructing the taxonomy, and generic product taxonomies are developed where possible. The consistency is not only limited to the product models in the same product line but across all the product lines.

A portion of NOV’s top drive taxonomy is presented in Figure 2. Key considerations in the top drive taxonomy are:

• All NOV top drive models in the product line adapt the same taxonomy, including hydraulic, electric, land-application and offshore-application top drives.

• An example of sub-units is provided in Figure 2, including drive motor assembly, guide dolly system, lubrication system, pipehandler and tool controls.

• Component/maintainable items in each listed sub-unit are fully captured.

Future Direction

NOV uses a master database to house and manage equipment taxonomy information. This database leverages the company’s current enterprise business application systems and is the foundation for improved failure reporting analysis and corrective action processes.

NOV has developed a structured approach and mechanism to drive consistency for how equipment maintenance learnings and failures are formally reported. With quality data and data analysis, the company is able to proactively identify equipment issues. With robust RCA, failure causes can be understood and positive preventative and corrective actions can be implemented and verified. DC

This article is based on a presentation at the 2017 IADC Asset Integrity & Reliability Conference, 22-23 August, Houston.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button