A Guide to AIM/Enduse Model
Go Hibino1, Rahul Pandey2, Yuzuru Matsuoka3, and Mikiko Kainuma4
1. AIM/Enduse: An Overview
1.1 Introduction
AIM/Enduse is a technology selection framework for analysis of country-level
policies related to greenhouse gas emissions mitigation and local air pollution control. It can also assist in energy policy analysis. It simulates flows of energy and
materials in an economy, from supply of primary energy and materials, through
conversion and supply of secondary energy and materials, to satisfaction of enduse
services. AIM/Enduse models these flows of energy and materials through detailed representation of technologies.
This guide is intended to enable the user to easily learn and work with
AIM/Enduse model. This chapter provides an overview of the model. Theoretical
formulation is explained in chapter 2. Methodology for data preparation, as explained in chapter 3, is written to guide the user in the process of collecting and
estimating data required for the model. Chapters 4 and 5 illustrate application of
the model to India and Japan. Thirteen appendices provide useful information like
terms of reference, GAMS code, description of AIM/Enduse database system, list
of international publications of relevant data, standard numbers for efficiency and
cost of selected technologies, standard numbers for calorific value, price and
emission coefficients of fuels, and classifications and data for AIM-India and
AIM-Japan.
1.2 Concept of the Model
AIM/Enduse is a bottom-up model of technology selection within a country’s energy-economy-environment system. Energy and material flows through technology systems in an economy, and consequent emissions, are modeled elaborately.
Selection of technologies takes place in a linear optimization framework where
system cost is minimized under several constraints like satisfaction of service demands, availability of energy and material supplies, and other system constraints.
1
Fuji Research Institute Corporation, Tokyo 101-8443, Japan
National Institute for Environmental Studies, Tsukuba 305-8506, Japan (on leave from
Indian Institute of Management, Lucknow 226013, India)
3 Kyoto University, Kyoto 606-8501, Japan
4 National Institute for Environmental Studies, Tsukuba 305-8506, Japan
2
248
Part IV. Manual: A Guide to AIM/Enduse Model
System cost includes fixed costs and operating costs of technologies, energy costs,
and other costs like taxes or subsidies. The model can perform calculations simultaneously for multiple years. Various scenarios including policy countermeasures
can be analyzed in AIM/Enduse. Key features of the model are described below
(refer to Appendix A for definitions of concepts and terms used in the model and
this manual).
1.2.1 Reference energy system
An integrated reference energy system is captured in the model. Energy and material flows through various technologies – from primary energy extraction, through
energy conversion and supply, to energy-enduse or final service conversion and
delivery – can be modeled in detail for a given year. Model’s structure is flexible
to permit modeling of energy and material flows and technology linkages to any
degree of scope.
As an aggregated option, a user can model energy and material flows beginning
from energy conversion technologies producing secondary energy from primary
energy, and ending at energy enduse technologies producing final services. Alternatively, as a disaggregated option, a user can model these flows beginning from
primary energy extraction technologies, passing through primary energy transportation technologies, energy conversion technologies, secondary energy transportation and distribution technologies, and ending at final service conversion and delivery technologies. In the disaggregated option, each technology can be further
represented as a combination of multiple devices, as explained in section 1.2.2.
Energy, materials and services can be represented in two ways in the model. An
energy or material that is supplied externally to the model is an external energy.
Similarly, a service that is demanded externally to the model is an external or final
service. Any energy or material that is produced and consumed within the model
is an internal service or energy (as shown in Figs. 1.2). Flow of an internal service
or energy is balanced within the model. This feature of AIM/Enduse allows extremely disaggregated modeling of reference energy system as described in previous paragraph.
Service demands are the main external drivers that trigger technology selection
decisions based on information on costs. Energy-mix and material-mix are derived
from the technology-mix. Finally, emissions of CO2, SO2, and NOx are derived
from the information on emission characteristics of energy, materials, and technologies.
1.2.2 Technology representation and selection
AIM/Enduse permits detailed representation of technologies where, on one hand, a
device and a combination of emission removal processes can be coupled together
(see Figs. 1.1 and 1.2), and on the other hand, multiple devices can be linked in a
complex network of sequential and parallel relationships that are close to reality
(see Fig. 1.3). Thus, in AIM/Enduse, concepts of device and removal process are
1. AIM/Enduse: An Overview
249
integral to the concept of technology, and need to be carefully understood by the
user.
A device may be defined as a distinct equipment or machine that is used in real
life. Alternatively, a device may be defined as an aggregate concept comprising
more than one distinct equipment or machines of real life (see Appendix A for
definition and Appendices J and L for examples in case of AIM-India and AIMJapan). A device can have multiple energy or material inputs and multiple outputs.
An example of device is smelting process in aluminum-making technology that
produces molten aluminum (internal service) from alumina (internal energy produced as internal service from Bayer’s process) by utilizing electricity (internal
energy produced as internal service from power plants) and fuel oil (internal energy produced as internal service from oil refineries). All devices are linked by a
complex network of flows of energy and materials within a reference energy system of an economy.
Additionally, attachments or retrofits of specific combinations of emission removal processes to regular devices can be represented in the model (see Figs. 1.1,
1.2 and 1.3). A combination of emission removal process, when attached to a device, removes a particular pollutant gas emission during the device’s operation.
AIM/Enduse permits attachment of SO2 and NOx emission removal processes at
three possible stages in a device’s operation – pre-combustion stage, in-situ combustion stage, and post-combustion stage. A combination of removal process is a
coupling of removal processes at these three stages. Coupling such a combination
with a device results in a combination of device and removal process, as shown in
Fig. 1.2. Coupling of a device with combination of removal process is optional in
AIM/Enduse. Ash-removal process introduced before a boiler, limestone addition
process during combustion of coal, and desulfurization of discharged flue gas, are
examples of pre-combustion, in-situ combustion, and post-combustion removal
processes respectively in a coal-fired power plant. Selection of a combination of
removal process (defined as exchange of one combination by another) for attachment to a device is decided based on its costs and emission removal performance.
For instance, under strict SO2 mitigation target, the model may select SO2 removal
processes as attachments to high SO2 emitting devices like high-sulfur coal-fired
power plants. Thus AIM/Enduse allows detailed technological assessment for both
regular devices and emission removal processes.
Two sets of decisions are made every year for technology selection in
AIM/Enduse – level of capacity recruitment and level of operation of each device
in a given year. Service demands determine the level of operating capacities required in a given year. Decision regarding recruitment-mix of devices in a given
year is made based on their annualized capital costs. Decision regarding operatingmix of devices in a given year is made based on their annual running cost which
comprises both energy costs and non-energy costs of operating the devices.
1.2.3 Stock transfers and multiple-year simulations
The model can be run simultaneously for multiple years as it simulates retirement
of devices and transfer of un-retired stock of devices across successive years. In-
250
Part IV. Manual: A Guide to AIM/Enduse Model
stalled capacity of a device in a given year is represented by its stock in the model
(see definition in Appendix A). Total stock of a device available in a given year is
the sum of un-retired stock transferred from previous year and new stock recruited
in the given year. Un-retired stock transferred from previous year is the total stock
available in previous year minus the stock that retired at the end of previous year.
A stock of a device retires after its life elapses from the year of its recruitment.
1.2.4 Outputs
Outputs of the model include recruited stocks of devices, operating quantities of
devices, use of energy-types, level of flow of internal energy or services, emission
quantities of CO2, SO2 and NOx, system cost including recruitment costs, operating costs of devices and costs of exchanging removal processes, in each year.
These outputs enable evaluation of a particular policy intervention or countermeasure on multiple criteria.
1.2.5 Scenario analyses and countermeasures
AIM/Enduse model permits analysis of various countermeasures. These are categorized as follows:
 Enduse stage countermeasures like efficiency improvement or better use and
management of devices.
 Emission tax on fuels for each of the following gases – CO2, SO2 and NOx.
 Energy tax on a (polluting) fuel to discourage its use.
 Regulatory constraint on quantity of emission of a gas in selected group of sectors in an economy.
 Regulatory constraint on quantity of use of an energy-type in selected group of
sectors in an economy.
 Subsidy on capital cost or operating cost of a device to promote its selection.
 Subsidy to promote attachment of an emission removal process to a device.
 Regulatory constraint on use of a clean or efficient device or its combination
with an emission removal process.
Combination of device and removal process
Energy
(Price;
CO2/SO2/NOx
emission factors)
Device
(Specific energy input; Specific service output;
Fixed cost; Operating cost; Life; SO2/NOx factor)
Combination of removal process
(Energy consumption factor; Fixed cost; Operating
cost; SO2/NOx removal; Retrofit factor)
Fig. 1.1. Energy-Technology-Service linkage AIM/Enduse model
Service
(Service
demand)
1. AIM/Enduse: An Overview
Device
Removal process
(pre-combustion)
Removal process
(in-situ-combustion)
251
Removal process
(post-combustion)
Combination of removal processes
Combination of device and removal processes
Fig. 1.2. Combination of device and removal processes in AIM/Enduse
Energy 1
Energy 2
Combination
of device and
removal
process 1
Internal
service/energy 1
Service 1
Combination
of device and
removal
process 2
Service 2
Energy 3
Internal
service/energy 2
Energy 4
Internal
service/energy 3
Combination
of device and
removal
process 2
Internal
service/energy 4
Combination
of device and
removal
process 4
Service 3
Internal service/energy 5
Fig. 1.3. Generalized technology representation in AIM/Enduse
1.3 AIM/Enduse Software
AIM/Enduse software comprises an integration of Optimization system (GAMS),
Database system (MS-Access), and Geographical information system (IDRISI)
(see Fig. 1.4). The mathematical formulation (see Chapter 2) is written and solved
in GAMS (see Appendix B). AIM/Enduse database system, developed using MSAccess, is the interface for GAMS program. It can supply input data for GAMS
program file AIM-CMB.gms and display results of simulation. It also provides a
user-friendly interface to the user for input of data, and design and analysis of scenarios or countermeasures (see Figs. 1.4 and 1.5). Interface with IDRISI (a GIS
and image processing software) permits geographical disaggregation and spatial
representation of input data and output results. This manual does not include description of IDRISI data requirement and operation.
Users input data in AIM/Enduse through MS-Access tables or forms that correspond to numbered databases in Fig. 1.5. The figure shows flow of information in
the database system. Combinations of indices indicated inside boxes for input data
and output of GAMS program highlight both extensive input data requirement and
252
Part IV. Manual: A Guide to AIM/Enduse Model
flexibility of output results in the model. For example, combination (L, J, K, M,
Y) in box ‘Device Specification’ means that required data for devices need to be
specified for every valid combination of device, service-type, energy-kind, gas,
and year. Similarly, combination (I, K, L, P, Y) in box ‘Energy Consumption’
means that energy use output can be specified for every valid combination of sector, energy-kind, combination of device and removal process, and year.
Appendix C describes the input data tables and each data item in detail. It also
describes the procedure for export of input data to GAMS file, running GAMS
program, import of output data from GAMS, and display of simulation results.
Simulation results can be viewed through a flexible interface that allows both tabular and graphical views of any combination of output variables. For instance,
year-wise results for energy use can be viewed by region, sector, or device; yearwise emission quantity of each gas can be viewed by region, sector, energy-type,
or device; recruitment costs, operating costs, costs of exchanging removal processes, taxes, and total system costs can be viewed by region, sector, or device;
etc. Users are advised to read Appendix C carefully to gain familiarity with the
model interface and its implementation.
User interface
User
Display of output
with pivot table and chart
Sum of Valu e En ergy_De vice
Remo val
C K1
C K2
COL BLR
Year
N ON
N ON
N ON
SFGD
_b as e
4 ,700 ,00 0
1 ,300 ,00 0
24 ,528 ,00 0
1 99 5
4 ,700 ,00 0
1 ,300 ,00 0
24 ,528 ,00 0
1 99 6
6 ,000 ,00 0
26 ,134 ,53 3
1 99 7
6 ,180 ,00 0
27 ,624 ,83 4
1 99 8
6 ,365 ,40 0
29 ,010 ,70 7
1 99 9
6 ,556 ,36 2
30 ,301 ,83 6
2 00 0
6 ,753 ,05 3
31 ,506 ,41 8
2 00 1
6 ,955 ,64 5
32 ,631 ,54 3
2 00 2
7 ,164 ,31 4
33 ,683 ,44 7
2 00 3
7 ,379 ,24 4
34 ,667 ,68 0
2 00 4
7 ,600 ,62 1
35 ,589 ,23 2
2 00 5
7 ,828 ,64 0
36 ,452 ,61 8
1 ,752 ,00 0
1 ,752 ,00 0
1 ,664 ,40 0
1 ,584 ,46 6
1 ,510 ,87 6
1 ,442 ,66 2
1 ,379 ,09 1
1 ,319 ,58 7
1 ,263 ,69 4
1 ,211 ,03 5
1 ,161 ,30 0
1 ,114 ,22 6
Import
module
AIM/Enduse
Database
Export
module
Export module
AIM/Enduse Database System (MS Access)
Input file of
GAMS Program
AIM/Enduse
GAMS program
Output file of
GAMS Program
AIM/Enduse GAMS Program (GAMS)
Input file
of GIS System
0.00
12.50
25.00
37.50
50.00
62.50
75.00
87.50
100.00
112.50
125.00
137.50
150.00
162.50
175.00
187.50
>=200.00
0.00
12.50
25.00
37.50
50.00
62.50
75.00
87.50
100.00
112.50
125.00
137.50
150.00
162.50
175.00
187.50
> =200.00
0.00
6.25
12.50
18.75
25.00
31.25
37.50
43.75
50.00
56.25
62.50
68.75
75.00
81.25
87.50
93.75
>=100.00
AIM/Enduse GIS System (IDRISI 32)
MS Access: Database management software designed by Microsoft Corporation
GAMS: General algebraic modeling system designed by GAMS Development Corporation
IDRISI 32: Geographical information system and image processing software designed
by Clark Labs, Clark University
Fig. 1.4. AIM/Enduse database system, Optimization system and GIS
1. AIM/Enduse: An Overview
253
Y : Year
1. Control Parameters
Emission Factor
(K , M, Y)
I : Sector / Region
Energy Price
(K, Y)
2. Region
J : Service-type
Service Demand
(I, J, Y)
3. Sector
K : Energy-type
4. Energy
L : Energy Device
5. Service
Device Specification*
(L, J, K, M, Y)
Combination of internal
energy and service (J, K)
Energy Consumption
(I, K, L, P, Y)
Stock in base year
(I, L, P)
Emission Quantity
(I, L, M, P, Y)
Maximum Share
(I, J, L, Y)
Service Supply
(I, J, L, P, Y)
P : Removal Process
6. Technology
Removal Specification*
(M, P)
M : Emission Gas
7. Combination
AIM/Emission (Country)
GAMS program
Technology Specification* L, J, K, M, P, Y)
Combination of device
(L, P)
Social Service Efficiency
(I, J, Y)
Stock Quantity
(I, L, P, Y)
Operating rate
(I, L, Y)
Operating Quantity
(I, L, P, Y)
Energy Efficiency
Improvement (I, K, L,
Y)
Subsidy Rate
(L, P, Y)
Recruitment Cost
(I, L, P, Y)
Energy / Emission Tax
(I, K, Y) / (I, M, Y)
Exchanging Cost
(I, L, P, P1, Y)
Energy / Emission Constraint
(K, ME, Y) / (M,
MQ, Y)
Input Data for GAMS
Tax Payment
(I, L, P)
8. Stock
9. Share
10. Social Service
11. Performance
12. Countermeasure
ME :Energy Constraint
MQ Emission Constraint
Database** of
Database System
Indices for GAMS
Operating Cost
(I, L, P, Y)
Output of GAMS
* Specification includes following data: fixed cost, operating cost, life, specific energy input,
specific service output, SO2/NOx emission coefficients
** Numbers before database names correspond to section numbers in Appendix C
Fig. 1.5. Information flow in AIM/Enduse database system
To summarize, salient features of AIM/Enduse are as follows:
 Technology selection using linear programming framework
 Representation of technologies as
 complex network of energy and material flows through multiple devices, and
 emission removal processes as options for retrofit attachments to devices
 Service demands as external drivers
 Technology selection based on annualized capital cost and running cost of
technologies including energy costs in a given year
 Retirement of technological stock at the end of its life, and transfer of nonretired stock across successive years
 Estimation of energy use for combustion and non-combustion operations of
technologies
 Estimation of quantity of CO2, SO2 and NOx emissions from fuel combustion as
well as non-combustion operations of technologies
 Integrated software comprising GAMS, MS-Access, and IDRISI
 User-friendly database system and interface to facilitate easy input of data, design of scenarios or countermeasures, and analysis of results.