Separate end-of-horizon approaches

[irm-codebase]

What can be improved?

End-of-horizon effects are one of the biggest concerns in pathways, so it's worth wile to test different approaches.

For now, I identify 3 approaches (at least 2 are missing here):

• No method: all CAPEX is assumed to happen at the year of installation.
• Salvage value: similar to no method, except that a salvage value is "repaid" at the end of the model run for assets that lived longer than the planning period.
• Annuity factors: you distribute costs according a technology discount rate, often called the weighted average cost of capital (WACC). This cost is then paid throughout the lifetime of a technology.

Goals

• Isolate costs for every investstep
• Implement discount factors in the objective function of the model
• Implement EoH mitigation approaches in different scenarios

Version

v0.1.0