Help with constructing user constraint (ZEV)

[cyang]

Hi,
I am trying to represent California's Zero Emission Vehicle (ZEV) mandate in a user constraint but I'm running into trouble. 

I thought the representation is relatively straightforward but I'm having trouble setting it up as a VEDA scenario file.

Each ZEV vehicle type has a weighting factor (a) that is multiplied by the number of vehicles sold in a given year to give more or less "credit" to that vehicle.  Full battery electric vehicles and fuel cells have greater weights than a plug in hybrid (which isn't a true ZEV).

The sum of these weighted ZEV sales has to be greater than or equal to a given percentage of total car sales.

The challenge I'm having is that I'm entering the weighting coefficients (a) for each individual ZEV vehicles as UC_NCAP values but I am also trying to sum across all vehicle sold (not just ZEVs), which has another coefficient (the ZEV Target %). 

Below is what I put together:
The first set of rows with "TCAR*" is meant to represent the 2nd half of the left side of the equation, the UC_NCAP values are the ZEV target percentages for each year.
All the subsequent rows are meant to just give the weighting factors for each type of vehicle.
 
but I know it's not quite right because when I view the UC Master, I can see that I have not differentiated between the TCAR* values (i.e. all the vehicles I'm trying to sum across) and the individual vehicle vehicle types (e.g. TCARELC1, TCARELC2, TCARFCH2, etc) and the coefficients that I'm specifying for the weighting factors is being written in the same spot as the Target percentage.

Hopefully this question is somewhat clear.  Thanks in advance for any help that you can provide.

Chris

~UC_Sets: R_E: AllRegions
~UC_Sets: T_E:
									~UC_T
UC_N	UC_ATTR~LHS	Pset_Set	Pset_PN	Pset_CO	Cset_SET	Cset_CN	Attribute	Year	LimType	UC_NCAP	UC_RHSRTS	UC_RHSRTS~0	UC_Desc
\I:  ZEV Mandate Overall Constraint
UC_ZEV_TZEV_to_2025	NCAP		TCAR*					2012	LO	-0.03	0	15
			TCAR*					2013		-0.03
			TCAR*					2014		-0.03
			TCAR*					2015		-0.06
			TCAR*					2016		-0.06
			TCAR*					2017		-0.06
			TCAR*					2018		-0.05
			TCAR*					2019		-0.07
			TCAR*					2020		-0.1
			TCAR*					2021		-0.12
			TCAR*					2022		-0.15
			TCAR*					2023		-0.17
			TCAR*					2024		-0.2
			TCAR*					2025		-0.22
			TCARELC1					2012		3
			TCARELC1					2017		3
			TCARELC1					2018		1.5
			TCARELC1					2025		1.5
			TCARELC2					2012		4
			TCARELC2					2017		4
			TCARELC2					2018		2.5
			TCARELC2					2025		2.5
			TCARFCH2					2012		9
			TCARFCH2					2017		9
			TCARFCH2					2018		4
			TCARFCH2					2025		4
			TCAR*PHEV10					2012		1.1
			TCAR*PHEV10					2017		1.1
			TCAR*PHEV10					2018		0.4
			TCAR*PHEV10					2025		0.4
			TCAR*PHEV30					2012		2.0
			TCAR*PHEV30					2017		2.0
			TCAR*PHEV30					2018		0.6
			TCAR*PHEV30					2025		0.6
			TCAR*PHEV40					2012		2.5
			TCAR*PHEV40					2017		2.5
			TCAR*PHEV40					2018		0.7
			TCAR*PHEV40					2025		0.7
			TCAR*PHEV60					2012		3.0
			TCAR*PHEV60					2017		3.0
			TCAR*PHEV60					2018		0.9
			TCAR*PHEV60					2025		0.9

[AKanudia]

what you have is equivalent to:

(weights x sales of targeted cars) >= target x sales of non-targeted cars

here is what it should be:

(weights x sales of targeted cars) >= target x (sales of targeted cars + sales of non-targeted cars)

rearranging, you will get new coeff for the targeted cars as (weight - target). you can leave the first part unchanged (TCAR*) as the values for targeted cars will get overwritten anyway.

hope the answer is somewhat clear.

[Antti-L]

 In this case it might be easier and more transparent to make use both sides, LHS and RHS, putting the target coefficients on the RHS and the weights on the LHS.  Then use UC_ATTR(NCAP,SYNC) on the RHS to activate both sides in the equation. 

[AKanudia]

That's a great idea Antti! So, any share UC could be defined more transparently like this, right?

I have been using the RHS only for GROWTH cases.

Thanks!

[Antti-L]

That's a great idea Antti! So, any share UC could be defined more transparently like this, right?

Right. Just define UC_ATTR~RHS with <type>,SYNC (here <type> was NCAP) and you will get both sides active.

[cyang]

Hi Antti and Amit,

Thanks so much for the input on the formulation. 
I understand Amit's point well enough to implement it in the way that he's suggested. 
I understand the concept of Antti's reply, but not the details well enough to implement it. 

I think that I'm supposed to add a column with:

UC_ATTR~RHS

NCAP, SYNC

and specify two columns for UC_NCAP

UC_NCAP~LHS

UC_NCAP~RHS

and then just move the Target values to the RHS (and changing the sign)

Is that correct?

thanks!

Chris

UC_N	UC_ATTR~LHS	UC_ATTR~RHS	Pset_Set	Pset_PN	Pset_CO	Cset_SET	Cset_CN	Attribute	Year	LimType	UC_NCAP~LHS	UC_NCAP~RHS	UC_RHSRTS	UC_RHSRTS~0
\I:  ZEV Mandate Overall Constraint
UC_ZEV_TZEV_to_2025	NCAP	NCAP, SYNC		TCAR*					2012	LO		0.03	0	15
				TCAR*					2013			0.03
				TCAR*					2014			0.03
				TCAR*					2015			0.06
				TCAR*					2016			0.06
				TCAR*					2017			0.06
				TCAR*					2018			0.05
				TCAR*					2019			0.07
				TCAR*					2020			0.10
				TCAR*					2021			0.12
				TCAR*					2022			0.15
				TCAR*					2023			0.17
				TCAR*					2024			0.20
				TCAR*					2025			0.22
				TCARELC1					2012		3
				TCARELC1					2017		3
				TCARELC1					2018		1.5
				TCARELC1					2025		1.5
				TCARELC2					2012		4
				TCARELC2					2017		4
				TCARELC2					2018		2.5
				TCARELC2					2025		2.5
				TCARFCH2					2012		9
				TCARFCH2					2017		9
				TCARFCH2					2018		4
				TCARFCH2					2025		4

[Antti-L]

Yes, correct.   

[cyang]

Antti and Amit

many thanks for your explanations and suggestions.  I have gotten the constraint working. 

Chris