Investment and operations of gas networks: A North African case study

Details

Supervisors

Lété, Quentin

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil en mathématiques appliquées, à finalité spécialisée

Abstract

enThis thesis develops and applies a stochastic gas network design model to the African Atlantic Gas Pipeline, a large-scale infrastructure project intended to connect Nigerian gas resources to Morocco through the West African Atlantic corridor. The objective is to analyze the investment logic of the project under demand uncertainty and to estimate the implicit valuation of gas demand required to justify different infrastructure choices. The model is formulated as a two-stage stochastic optimization problem, where first-stage decisions represent infrastructure investments and second-stage decisions represent scenario-dependent operations. The formulation includes nonlinear gas transmission physics through Weymouth-type pressure-flow constraints and distinguishes between passive pipeline arcs and active compressor arcs. It is then extended to include Liquefied Natural Gas (LNG) as an alternative transport option, with liquefaction capacity, regasification capacity and maritime shipping costs. A welfare-based formulation is finally introduced in order to move beyond pure cost minimization. Instead of assuming that all demand must be served, the model associates served demand with a concave benefit function depending on a valuation parameter. This makes it possible to address the central research question of the thesis: what is the implicit valuation of gas that justifies investment in the African Atlantic Gas Pipeline ? The numerical case study shows that, under fixed demand, the baseline solution selects a complete pipeline corridor from Nigeria to Morocco, with an annualized system cost of approximately 4.1 B€/year. LNG remains almost unused under pure cost minimization, but becomes relevant in some welfare-based scenarios. The welfare analysis shows that the implicit valuation of gas cannot be reduced to a single break-even value. Under the total welfare specification, the first corridor investments appear at V=307 M€/bcm, the full Nigeria--Dakhla corridor becomes optimal at approximately V=1 B€/bcm, and full demand satisfaction requires V=3544 M€/bcm. These results highlight that the economic justification of the project depends on the scope of the welfare function and on the level of infrastructure ambition considered.