Mohammad Ahanchian 2/4/2014

Like

1992 Views

Dear friends,
I have tried to model phasing out old vehicles and replacement with a new vehicle having different attributes. I have a bottom-up analysis, and I need to phase-out a certain amount of vehicles in each year (the data for each year are available and should be modeled with exact value). I have 67100 old vehicles in stock, and defined the sales as bellow:

For old vehicle: Interp(2010, 0, 2013, 0, 2014, -23978, 2015, -19578, 2016, -50, 2017, -7777, 2018, -5612, 2019, -703, 2020, 0, 2021, 0, 2022, 0, 2023, -7254, 2024, -2139, 2025, 0, 2026, 0, 2027, 0, 2028, 0, 2029, 0, 2030, 0, 2031, 0, 2032, 0, 2033, 0, 2034, 0, 2035, 0, 2036, 0, 2037, 0, 2038, 0, 2039, 0, 2040, 0)

For new vehicle: Interp(2010, 0, 2013, 0, 2014, 23978, 2015, 19578, 2016, 50, 2017, 7777, 2018, 5612, 2019, 703, 2020, 0, 2021, 0, 2022, 0, 2023, 7254, 2024, 2139, 2025, 200, 2026, 200)

The problem is that, all the energy and environmental results have minus sign!!!! I think it occurs because of minus in the sales!!!

I tried using retirement lifecycle but i see that its not very accurate.
Please help me for this model.
Tnx

Taylor Binnington 2/4/2014
 

Like

1990 Views

Hi Mohamad,

You're correct, in that you will see negative environmental loadings if you enter negative sales data. LEAP does not expect to see negative numbers of sales, and so it treats it as any other number - this isn't a good way to go about entering vehicle retirements.

I'm not completely certain what you're trying to do, but for planned vehicle retirements, I think that building a lifecycle profile to mimic these retirements would be the best way to do it. The simplest case that I can think of would be to define your existing stock of 'old' vehicles using the base year Sales variable (in Current Accounts), rather than the Stock variable. This way, all of your initial stock will be of the same vintage: zero years old. This will make your retirement calculation easier. Make sure that in your scenarios, you reset this Sales variable to zero though, indicating that no 'old' vehicles are sold in future years.

You'll then need to create a Survival Profile (defined as a lifecycle profile) to control the retirement of these vehicles. Each value in this lifecycle profile describes the percentage of vehicles that survive after a number of years. Since all your vehicles will be of the same age, though, there will be a direct correspondence between the values entered in this profile and the remaining number of vehicles.

If you haven't seen it, a quick read through our Transport Analysis help page will help provide some background:

http://www.energycommunity.org/WebHelpPro/Demand/Vintaging_Calculations.htm

Note that forced retirements, in a transport analysis, is something that is on our 'to do' list. We'd like to implement a simpler way to do this, but until then, this is a good way to work around the issue.

Hope this helps,

Taylor

Mohammad Ahanchian 2/4/2014
 

Like

1 Like 1972 Views

Dear Taylor,
Thank you for your comprehensive explanation.
I am going to model replacing the old vehicles with a brand new one. Therefore, for each vehicle that phases out, a new vehicle enters to fleet and the total number of this fleet (old vehicle and newly introduced vehicles) should remain always same in each year.

Lifecycle profiles are defined to be either exponential or data point, the latter one only accepts percentage! If I have a good understanding of life cycles, it might not be applicable to control the retirements for this case.

Yeah I fixed the stock, and define variables for sale in Current Account.

Oops! So I wish that I this version had forced retirement.

Consequently, I couldn't use lifecycle for this kind of constraints. However, I used another method, and it worked with negligible error!

I defined lifecycle profiles to have almost negligible effect on the retirement and then added some more vehicles, then breakdown the results in excel and omitted the outliers!

I have a comment on lifecycle profiles, since the results are very sensitive to the inputs, it would be a good idea to provide sensitivity analysis in LEAP.
Thanks again.

Taylor Binnington 2/6/2014
 

Like

1971 Views

Hi Mohamad,

Yes, from the sound of it, a forced retirement option would be very useful to you. Though I do think that what you're trying to do is a little bit non-standard, from a policy perspective (i.e. substituting old for new cars, no matter that age of the old car, while maintaining the same total number of vehicles) but in the meantime, I'm glad to hear that you've worked around it.

I'm not quite certain why the Survival Profile route will not work for you though. Since no more 'old' cards are being sold, you know the total stock of 'old' cars, which means that you can retire a fixed number of them each year by inputting the appropriate percentage into the Survival Profile that you assign to these cars.

I'm also curious - can you describe what you mean when you say that results are very sensitive to the inputs in lifecycle profiles? Off hand, I don't think there should be any unusual behaviour, but perhaps I'm misunderstanding you?

Thanks very much, and good luck,

Taylor