A generalized MILP framework for plant-level decarbonization

Energy-intensive plants must navigate technology, fuel-price, and policy uncertainty when selecting least-cost decarbonization pathways. We present a plug-and-play, two-stage MILP framework that jointly optimizes capacity expansion and commissioning schedules, hourly site-energy operation, and an emissions-allowance procurement and intertemporal-banking strategy. To our knowledge, this is the first framework to co-optimize multi-technology investment timing with hourly operations under an explicit carbon-trading regime with banking, within a single scenario-based model. The formulation integrates (i) multi-decadal capital planning with hourly dispatch, (ii) an allowance-account module that computes ETS-consistent compliance costs with banking, (iii) correlated trajectories for grid-carbon intensity, fuel and power prices, and ETS design, and (iv) CAPEX learning curves for electrification, hydrogen, CCS, and biomass. Applied to a UK epoxy-resin facility (2025–2055) and stress-tested across five market–policy scenarios, four robust insights emerge. First, electrification is least-cost/lowest-emissions only if grid intensity falls below 50 g CO2 kWh-1 by 2035; a slower trajectory adds up to 745 kt CO2 over project life. Second, a carbon-price corridor alone cannot bridge the green-fuel premium; long-dated hedges (PPAs, forward/swap positions, collars) are needed to reward first movers. Third, permitting credit banking delivers an additional 110 kt CO2 of cost-effective abatement by rewarding early action and sharpening the cost–emissions Pareto frontier; banning banking flattens that frontier and raises compliance cost. Fourth, CCS and biomass remain uneconomic unless negative-emission credits are fully tradable within the ETS. The framework is transferable and provides a practical, scenario-driven stress-testing tool for regulators and operators seeking deep industrial decarbonization.