Assets which expire between milestone years show `decommission_year` as next milestone year

[alexdewar]

I think this is incorrect. It can lead to a weird situation where it looks like assets have lived beyond their lifetime, when really they've essentially been decommissioned before the next milestone year has begun. Let's fix this to make the output more consistent for users.

One way we could implement this is to change Asset::decommission() so that it will use either the asset's max_decommission_year() or the year argument, depending on which is lower.

[tsmbland]

@ahawkes

Minor point, but we have a potential scenario where the decommission year of an asset falls between milestone years.

For example, the commission year is 2020 and the lifetime is 15 years, meaning an expected decommission year of 2035. If the time framework is [2020, 2030, 2040], then we'll decommission it when running the model for 2040 (assuming it doesn't get decommissioned early). In this case, what should we report as the decommission year for that asset in the output data: 2040 or 2035?

Apologies if this has been asked already

[tsmbland]

@ahawkes

Minor point, but we have a potential scenario where the decommission year of an asset falls between milestone years.

For example, the commission year is 2020 and the lifetime is 15 years, meaning an expected decommission year of 2035. If the time framework is [2020, 2030, 2040], then we'll decommission it when running the model for 2040 (assuming it doesn't get decommissioned early). In this case, what should we report as the decommission year for that asset in the output data: 2040 or 2035?

Apologies if this has been asked already

Alternatively, in this scenario, the agent could notice in 2030 that the lifetime of the asset is not sufficient to reach the next milestone year, so could decommission it early to replace with a new asset that fully covers the 2030-2040 period

[ahawkes]

This is a likely scenario, and I think tricky to handle well. Suggest for now we record decommission year as 2035. But in reality, I guess implicitly we are assuming it was replaced by the same thing, for the period 2035-2040. In future if we want to calculate total investment cost, we'd need to factor this in - i.e. the annualised capital cost would continue to 2040 - otherwise we'd have a magical 5-year gap that was free of capital cost but somehow still had the asset operating.

…

________________________________ From: Tom Bland ***@***.***> Sent: Wednesday, October 08, 2025 14:49 To: EnergySystemsModellingLab/MUSE_2.0 ***@***.***> Cc: Hawkes, Adam D ***@***.***>; Mention ***@***.***> Subject: Re: [EnergySystemsModellingLab/MUSE_2.0] Assets which expire between milestone years show `decommission_year` as next milestone year (Issue #867) [https://avatars.githubusercontent.com/u/23723407?s=20&v=4]tsmbland left a comment (EnergySystemsModellingLab/MUSE2#867)<#867 (comment)> @ahawkes<https://github.com/ahawkes> Minor point, but we have a potential scenario where the decommission year of an asset falls between milestone years. For example, the commission year is 2020 and the lifetime is 15 years, meaning an expected decommission year of 2035. If the time framework is [2020, 2030, 2040], then we'll decommission it when running the model for 2040 (assuming it doesn't get decommissioned early). In this case, what should we report as the decommission year for that asset in the output data: 2040 or 2035? Apologies if this has been asked already Alternatively, in this scenario, the agent could notice in 2030 that the lifetime of the asset is not sufficient to reach the next milestone year, so could decommission it early to replace with a new asset that fully covers the 2030-2040 period — Reply to this email directly, view it on GitHub<#867 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AC37JLN3VGU37EZK3TM7IJT3WUI43AVCNFSM6AAAAACHZVSE5WVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGOBRGY2DCMRWHA>. You are receiving this because you were mentioned.Message ID: ***@***.***>