Decommissioning Optimization in a Multioperator Landscape

Fig. 1—Relative overall decommissioning costs of several scenarios (midcase input).

You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.

To ensure continued access to JPT's content, please Sign In, JOIN SPE, or Subscribe to JPT

Total offshore decommissioning expenditures (ABEX) for the Netherlands are significant, currently estimated at more than EUR 5 billion. This equals one-third of the total future offshore revenues. Energie Beheer Nederland (EBN), the state participant in all offshore gas assets, is responsible for approximately 40% of decommissioning costs. EBN is investigating opportunities to reduce these costs without compromising safety or environmental standards. The optimization model presented in the complete paper is the first multioperator offshore network-optimization model that considers decommissioning in the Netherlands.

Introduction

A network-optimization model including platforms, wells, and technical production forecasts was built to investigate value optimization on a portfolio basis, considering gas revenues, operating expenses (OPEX), and ABEX. Future offshore decommissioning projects are shared across the coming 3 decades by 11 operators. The resulting complexities can be overcome in several ways, but these, in turn, result in organizational challenges. The complete paper thus discusses five options for decommissioning planning with different collaboration strategies, which are defined as scenarios:

  • Reference scenario—no collaboration. Stop production at zero economic margin.
  • Operator optimization—no collaboration. Optimize within operator portfolios.
  • Cross-operator optimization—all operators collaborate to full efficiency.   
  • Pairs—create a pair of operators that cooperate bilaterally.
  • Decommissioning company—a separate company executes all decommissioning activities.

Note that the reference scenario does not include any cost reduction or ABEX-optimization strategies. It calculates the timing and amount of realized ABEX, which is effectively the amount of estimated ABEX per individual project.

Model Overview

The model used for this study combines net-present-value (NPV) -optimization strategies while describing a real network of wells, platforms, and pipelines. The technical production forecasts of gas fields are coupled by wells to platforms. Existing infrastructure connects platforms and the onshore delivery points. OPEX has to be carried by the revenues. Current OPEX and tariffs are used and assumed to be constant over time. This might cause platforms to have a cessation-of-production date that is slightly earlier than in reality, because it is usually assumed that OPEX decreases in late life because of OPEX-reduction measures. The ABEX input used is generic; an average value is used for well-decommissioning and platform-decommissioning costs. A distinction is made between small and large platforms.

Model. A mixed-integer-programming (MIP) optimization model was developed. MIP models describe difficult problems and require advanced solvers. The model is a multiperiod network investment model. The model decides when the abandonment projects are executed such that NPV is maximized while satisfying a set of legal, technical, physical, and economic constraints.

Input. The simulated network of wells, platforms, and pipelines resembles the actual Dutch offshore gas system. The assumed OPEX per platform and technical production forecasts per field date from December 2016. Generic platform- and well-decommissioning costs are used. The model input consists of the following:

  • The full Dutch offshore gas network
  • Technical gas-production forecasts
  • 504 platform wells
  • 86 small platforms
  • 57 large platforms
  • Fixed and variable OPEX
  • 11 operators
  • Gas production from subsea completions being tied to the connected platform

The number of wells and platforms represents the Dutch offshore system but excludes any condensate and oil production, oil platforms, oil wells, subsea wells, and standalone exploration wells (only platform wells are included), although any oil-associated gas that flows through a gas platform is included in the modeled gas stream.

As a reference scenario, midcase costs are evaluated, but without any optimization strategies. This will result in the total ABEX that should be reserved for all decommissioning projects. Two classes of platforms are considered, small and large. The classification is based on the weight. A topside that weighs more than 1200 t, typically a limit for sheer-leg vessels in the southern North Sea, is assumed to be a large platform. Only one type of well is considered, assuming that plug-and-abandonment (P&A) time and costs will average out. All costs that are estimated are verified with experts. Nevertheless, sensitivity runs on these estimations were performed to understand if the model is sensitive to certain cost assumptions.

Model Decisions. The life cycle of a platform is simplified in the model, which assumes that every platform has four possible stages: production, P&A, lighthouse, and decommissioning. In the initial stage, production mode, it is considered that a platform has a fixed and variable OPEX and possible revenues from produced gas to cover the OPEX. If a platform is still economically viable as a network node, it stays in production mode. A platform ceases production either by cessation of production or because no more gas is produced or flows over the platform. The model is constrained to decommission the platform within a set time frame following cessation of production. The general platform life-cycle assumptions include the following:

  • The production is limited by a technical forecast; the model determines economic optimum (NPV).
  • The model timesteps are on a yearly basis, because the forecast is on a yearly basis.
  • The wells of a platform are all plugged and abandoned in the year following cessation of production.
  • Lighthouse mode and decommissioning mode require all subject wells to be plugged and abandoned.
  • Lighthouse mode is optional; the model decides what is economically optimal.
  • Decommissioning must be executed not later than 5 years after cessation of production, which can be extended to a maximum of 8 years in lighthouse mode.

To understand the added value of different strategies, the complete paper regards the five scenarios described previously. The general challenge for these scenarios is to find a tradeoff between ABEX reduction and organizational complexities, as a result of more optimization and collaboration. If one enlarges the platform pool to allow for more optimization, will one see more benefit? And, if so, how much and is it worth the effort?

Results

Midcase Results. The reference scenario is taken as the default scenario to evaluate the effectiveness of an optimization strategy. Fig. 1 above shows the results of four scenarios. Compared with the reference scenario, the operator-optimization scenario results in an overall decommissioning cost of 73%, while cross-operator optimization leads to an overall total decommissioning cost of 69%. Note that the step from optimization within the portfolio of each operator to optimization across operators results in an additional reduction of only 4%. This, however, represents a cost reduction of almost EUR 230 million in the midcase. It should be noted that, in the Netherlands, the operators are either relatively large, and thus can ­easily optimize within their own portfolio, or they are relatively small. The smaller operators benefit the most from cross-operator optimization in relative terms. However, every operator benefits in an absolute sense from full collaboration, because a small percentage of cost reduction of a large portfolio still represents significant value.

Fig. 1 also illustrates that the decommissioning-company scenario results in the lowest overall decommissioning cost. While both optimization scenarios have clusters of jobs for which new contracts have to be negotiated and in which learnings are not shared (or kept, because there is idle time between campaigns), the latter scenario is defined on the principle of continuous learning and long-term contracts through a steady activity rate. This results in an additional cost reduction of 15% compared with the operator-optimization case, which represents almost EUR 900 million.

Pairs. The model allows for grouping of operators to explore the potential mutual benefit of their collaboration. It assumes in this case that each operator optimizes within its portfolio, but two operators also fully collaborate with one another (bilaterally). This option was not tested extensively; therefore, only some indicative results are given in the complete paper.

Sensitivity Runs. The midcase input parameters have been verified by a selected number of operators. However, most input parameters may vary widely in reality. Variations in important input parameters could very well change results and general conclusions. Therefore, several runs were performed to check if the model is sensitive to certain parameters. Sensitivities were checked by running low- and high-case values for a selection of parameters. The results indicate that the general conclusions for the midcase input are not very sensitive to changing the most-important model parameters.

Conclusions

A network model was built to optimize decommissioning planning and to quantify decommissioning expenditures. Several decommissioning-optimization strategies were investigated. Each optimization scenario results in cost reductions. However, clear differences appeared between the scenarios.

Careful planning of each separate operator decommissioning portfolio results in large overall cost reductions compared with the reference scenario. However, mainly large operators, with large portfolios, benefit from optimizing within their own portfolio. Collaboration in a selected operator pair represents additional value; although it is less significant than full cross-operator collaboration, it is more likely in a practical sense and much easier to achieve. Cross-operator collaboration results in significant cost reduction but requires 11 operators to collaborate fully. In any case, to optimize the same cross-operator portfolio, a dedicated decommissioning company results in far larger cost reductions. Establishing a dedicated decommissioning company requires a full reorganization of the decommissioning strategy, but the cost reductions are very significant—potentially more than 40%.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 186147, “Decommissioning Optimization in a Multioperator Landscape,” by T.D. Huijskes, R.E. Stoeller, E. Kreft, E.T. van Ewijk, C.T.J. van Langen, and B.C. Scheffers, Energie Beheer Nederland, and G.W. de Mare, H.C.M. Bossers, M.H. de Wolff, and G.G. de Vries, Ortec, prepared for the 2017 SPE Offshore Europe Conference and Exhibition, Aberdeen, 5–8 September. The paper has not been peer reviewed.

Decommissioning Optimization in a Multioperator Landscape

01 January 2018

Volume: 70 | Issue: 1

STAY CONNECTED

Don't miss the latest content delivered to your email box weekly. Sign up for the JPT newsletter.