API Bull 1178-2017 pdf free download

API Bull 1178-2017 pdf free download.Integrity Data Management and Integration.
1 This bulletin provides a compendium of methodologies and considerations for integrating the underlying data used to support integrity m anagement. Any one approach, let alone the entirety of the document, may not be appropriate or applicable in all circumstances. The document reviews possible approaches for consideration by operators in the context of their specific circumstances. The primary focus of this bulletin is the m ethodologies and processes used to s patially integrate and normalize the data to support the application of comparative techniques used in interpreting integrity data, with par ticular em phasis on in-line i nspection ( ILI) data. T he doc ument b egins with a di scussion of general dat a-quality processes, goal s, an d c onsiderations s uch t hat da ta quality approaches c an b e considered in the context of the data integration processes. An impediment to informed integrity decisions is the inability to efficiently review a broad spectrum of data in a format that has been normalized and spatially aligned. With the variations in or ganizational structures, integrity management programs, and technologies used across the pipeline sector, individual operators design data integration procedures that are customized to their organizational structure, processes, and pipeline systems. Properly m anaged and i ntegrated dat a s upport a gile anal ytics t o i ntegrate ne w d ata as they bec ome available an d t o r ecognize c oincident e vents and p atterns. T he dat a s ource m ay b e f rom w ithin an organization, or may be external to the company, as in the case of representative data based on industry experience or m anufacturing processes. The intent is to em power o perators to efficiently a nalyze an d integrate threat- and integrity-related data to support their integrity management programs.
4 Managing pipeline integrity data historically involved the rather manual process of populating data within spreadsheets or disparate databases. Transitioning to an enterprise database to manage large pipeline integrity data sets provides an operator with several advantages, including the following: — Improved auditing and traceability: When spreadsheets are created, the logic and judgment that is applied while an individual is manipulating data is not captured, or easily understood. In most cases, this logic exists only in the mind of the individual who created the spreadsheet, which may result in compliance risk. — Improved t racking of dat a c orrections: P ropagating corrections t o da ta er rors ac ross multiple dependent s preadsheets, or bac k t o t he original data s ources, i s difficult a nd may po tentially introduce further errors. — Improved s afeguards against h uman er ror: H uman er rors, s uch as versioning er rors a nd corruption errors, can compromise the i ntegrity of da ta entry. D atabases and t heir as sociated graphical interfaces facilitate the implementation of quality rules and constraints that mitigate the potential for human error. — Improved r esource utilization: D atabases m ay provide i mproved ef ficiency o ver dat a management that uses disparate spreadsheets.
— Improved data security: Server-based data may be more difficult to access and propagate than individual files, which can be easily transferred to local or portable drives. — Improved s calability: Spreadsheets and s mall-scale dat abases m ay ha ve s ize l imitations t hat enterprise databases do not have. Data Quality Oversight 5 General 5.1 The as pects of data qu ality listed in 5.2.3 are examples of el ements an oper ator m ay consider when developing a data m anagement system, but the extent to which they are relevant varies d epending on asset complexity a nd organizational s tructure. As with an y s ystem, continuous improvement i s a c ore principal. T he s ystem i n this context is often referred to as a geographic i nformation s ystem ( GIS) or database, but i t m ay be a compilation of app lications and databases, with a map-based visualization being just one aspect of the solution.