Improved Production Forecasting for Work Programme Execution Performance

Abstract

Reliable production forecasting sits at the heart of good asset management and business planning in the upstream oil and gas sector. Decline Curve Analysis has served as the standard forecasting tool for planning purposes mainly because it is straightforward to apply and does not demand large volumes of data. In practice, a production forecast is pulled in many directions at once: processing facilities impose constraints, well interventions alter rate trajectories, unplanned shutdowns create deferments, and the work programme itself changes shape as the year unfolds. When the forecast is not continuously refreshed to reflect these realities, the gap between what was predicted and what was actually produced tends to widen quickly, eroding the confidence of planners, management, and shareholders alike. This paper describes an integrated production forecasting workflow developed by the Newcross Exploration and Production team to close that gap. The central idea is straightforward: every element of the asset work programme that has a bearing on production should be captured in the forecast input deck, and that deck should be kept current as execution progresses. Achieving this in practice required a structured data management approach, a defined review cadence, and a direct link to the RADA Asset Management Solution (RADA AMS), which provided the data infrastructure needed to make continuous updates practical rather than aspirational. When this workflow was applied alongside Well and Reservoir Facility Management (WRFM) best practices, the alignment between forecast and actual production improved markedly. Planned activities were scheduled more efficiently, execution risks were caught early enough to act on, and resources could be redirected in a timely way. The paper explains how the workflow was designed and why each element matters, then presents the field results and draws out the wider lessons for asset teams considering a similar approach.