Proper well Plug & Abandonment (P&A) design requires an accurate picture of the current well status, since it represents the starting point to ensure the permanent isolation of zones with potential flow.
The greater the amount and quality of data available, the clearer the picture of the well, the better the P&A design. Then, the selection of the most suitable technologies for the specific case allows optimizing the operational sequence thus minimizing the related costs, typically around 50% of the global asset decommissioning cost.
Missing or inaccurate data at the P&A design phase results in a conservative approach, due to the many uncertainties regarding the well.
Data quantity and, mainly, quality is often dependent upon the age of assets and the existence of a proper Well Data Management System. This paper brings the P&A experience of a platform-based six-well asset in Africa, with no available well data repository resulting in a thorough "hunt for information" to generate a small database and identify the main data gaps relevant to the P&A operation being designed.
The collected data, ranging from 10 to 45 years of age, was highly fragmented and heterogeneous. For some wells, the missing information even included well construction files, cement logs, some overburden lithology parameters and, for one of the wells, the characteristics of wellhead and annulus fluid.
The execution of an uncertainty analysis based on the good practices applied in the North Sea and the Gulf of Mexico, showed that P&A cost could have varied between M$ 3.7 and 7.5 per well. The latter value was due to additional well barriers resulting from the uncertainty of well status.
Analysis of the initial BoD and potential optimization of the P&A sequence resulted in the definition of three scenarios: Best Case (full optimization), Worst Case (no optimization) and Base Case (limited optimization based on gathered information).
Two wells are discussed in this paper: Well A, with a moderate level of information, translated into an increase from the Best Case of 18%, representing around M$ 0.7; Well B, with very poor data available, translated into an increase from the Best Case of 75%, of which around 36% or M$ 1.3 related to missing information. As a result, the final P&A scenario led, for the entire asset, to an estimated increase of M$ 6 to 7 vs. the Best-Case scenario.
A simple and adequate Well Data Management System, also described in this paper, would have allowed to provide a cost-effective P&A design. The above "M$ 6 to 7" value (around 1 M$ per well) can be considered as a good indication of the value of such a system, although it is conservative in nature, since further benefits could have been also achieved during the production life of the asset.