Does HOPE support multi-stage GTEP

[Kaisan-Li]

Description

Hi,
Thanks for your great work.

I wonder if HOPE currently supports multi-stage GTEP for long-term planning (e.g., 2025-2060, 5 years per stage). Could you give me some tips?

[swang22]

Hi Kaisan,

Thanks very much for your question and for your interest in HOPE.

The GTEP mode supports generation and transmission expansion planning with either full-year hourly chronology or representative-day formulations. However, the current public version does not appear to natively support a multi-stage / multi-period long-term expansion formulation in the sense of sequential planning stages (for example, 2025–2030–2035–...–2060) with explicit intertemporal linking of build decisions across stages.

At the moment, GTEP setup is more like a single planning horizon model:

• investment costs are formulated in annualized form,
• operations are modeled over one representative study year,
• reserve/policy settings are defined for that case horizon,
• and the public input structure does not seem to expose stage-indexed build vintages, retirement carryover, or inter-stage state transitions.

However, there are a few practical ways to you can manually to do multi-stage long-term pathways with HOPE:

1. Rolling-horizon / sequential runs

   • Run HOPE for 2025, then update the system to reflect selected builds/retirements.
   • Use that updated fleet/network as the input for 2030, and continue stage by stage.
   • This is probably the most practical workaround with the current version.

2. User-prepared milestone cases

   • Create separate cases for 2025, 2030, 2035, etc.
   • For each milestone, update demand, fuel costs, policy assumptions, and candidate technology costs externally.
   • Then compare the resulting expansion paths across milestones.

3. Prototype extension to native multi-stage GTEP

   • In principle, HOPE could be extended by adding a planning-stage index to investment variables and linking capacity evolution across stages.
   • This would also require stage-specific demand/policy/cost inputs and discounting across stages.

If your goal is a 2025–2060 study with 5-year steps, I would currently recommend the first approach: a sequential workflow with carefully updated input data between stages.

I think we should add this as a future feature request / roadmap item for HOPE. A native multi-stage formulation would be a valuable extension for future development.

Thanks again!
Best,
Shen