Best Practices for Successful Implementation of Integrated Asset Model Based Well and Reservoir Workflow Automation - A Practical Learning Experience from Mature Brown Fields

Abstract

Although asset model based workflows are not new in digital oil field, drive towards automation is more and more increasing to move from periodic to continuous optimization for improving process and operational efficiencies. Integrated asset model forms the basis for number of well & reservoir workflows and can aid in standardization and automation. Models are heart of number of activities such as surveillance, calibration, optimization and forecasting within asset. These activities are inherent in most of workflows performed by engineers on any project or task. This paper is intended to discuss the best practices based on lessons learned from implementations in large mature brown fields in ADCO where sustaining allowable production, well performance issues, Production reconciliation, facilities bottle-necks & real-time data availability were major challenges.

The corporate asset strategy shall have a vision towards automation and its benefits to organization's strategic objectives. Workflow automation for an asset will depend greatly depends on the objectives from a business process to accomplish and should be bringing maximum value. This must result in tangible impact whilst providing means to start establishing a new mindset. The initial efforts must focus on ‘fixing the basics’ such as mapping of existing detailed workflow steps of a process, identify key data required, thorough gap-analysis, improve data reporting & QA and agree on common definitions before automation takes place. Expectation setting with stakeholders should be done early in process and operations staff need to be involved early to help establish objectives and ensure workflows are adequate to their use. Prototype and phased workflow deployment approach shall be adopted. Engineers need to be given a chance to develop to trust automated system before workflows can be fully automated.

Improving just process efficiency should not be end of goal of automation however engineers should be able to identify optimization opportunities in quick time. Automated model calibration can pinpoint data of poor quality and justify its improvement. Exception based well & facilities network surveillance is a common feature of automation hence rate estimation what if methodologies, validation limits, exception handling, pressure drop thresholds & pre-configuration of multiple operations scenarios shall be thoroughly discussed. Historical data trending in workflows can support decision making and add a value. Workflow and model governance need to be managed efficiently for automation to survive. Coherent and effective management information such as rolling-up of production volumes, deferment, operations KPIs need to be reported as a result of this automation to increase transparency. Agility, scalability and interoperability are key factors and must be supported by automation system.

The authors primarily discuss challenges addressed in workflows deployment, data integration & improvements, capability development and change management mechanism in these implementations.