Energy models: Methods and characteristics Najmeh Neshat Mohammad Reza Amin-Naseri Farzaneh Danesh Department of Industrial Engineering, Tarbiat Modares University, Tehran, Iran Abstract economic growth and enhancing welfare and Given the importance of models in complicated equality in the long term; (3) the energy-environ- problem solving, an inappropriate energy model ment interactions include the impacts of energy pol- can lead to inaccurate decisions and poor policy icy on environmental phenomena such as climate prescriptions. This paper aims at developing a deci- change, resources stocks, ecosystem and human sion support tool with which the selection of appro- health. Therefore, the major difficulty of the EPM priate model characteristics can be facilitated for not only lies in its multi scales aspects (temporal and developing countries. Hence, it provides a compar- geographical), but also in the necessity to take into ative overview of different ways of energy models account the economic, technical, environmental characterization and extracts the underlying rela- and social criteria. Modelling of complex problems tionships amongst them. Moreover, evolution of can lead to better decisions by providing decision dynamic characteristics of energy models for devel- makers with more information about the possible oping countries is identified according to the previ- consequences of their choices. Hence, energy mod- ous studies on the developed and developing coun- els as valuable tools for dealing with complicated tries. To do this, it reviews the related literature and problems can help decision makers to overcome follows a systematic comparative approach to this difficulty. Energy models are useful mathemati- achieve its purposes. These findings are helpful in cal tools based on the system approach and the best cases where model developers themselves are look- model should be determined based on the problem ing for appropriate characteristics in terms of certain that decision makers endeavour to solve. purpose or situation. Recently, the total number of developed energy models has grown tremendously and they vary con- Keywords: energy models; energy models charac- siderably in characteristics and features. Hence, the terization; developing countries key question is that ‘which model(s) and character- istics are most suited for a certain purpose and situ- ation?’ Also ‘what are the underlying relationships amongst these characteristics?’ Given the diversity of the possible characterization approaches, this 1. Introduction study aims at developing a comparative picture of Over the past three decades, energy planning and them which can provide insight not only in the dif- management (EPM) has played an essential role in ferences and similarities between them but also in long-term social, environmental, and economic pol- the underlying relationships amongst them. icy making of countries. Energy systems as an inte- Moreover, the evolution of dynamic characteristics gral part of socio-economic systems of societies of energy models for developing countries is identi- have several cross-disciplinary interactions with fied according to the previous studies on the devel- economy, society, and environment. To name a oped and developing countries. To do this, it few, (1) the energy-economy interactions consist of reviews the related literature and follows a system- changes over time in price elasticity of demand and atic comparative approach to achieve its purposes. also impacts of macroeconomic activity on energy In this paper, we will first give an introduction to demand; (2) the energy-society interactions are the the different energy model categorization approach- impacts of energy cost on labour productivity, capi- es as section 2 and then indicate their relationships tal formation, energy consumption and therefore, via a schematic diagram. Trends in dynamic char- Journal of Energy in Southern Africa • Vol 25 No 4 • November 2014 101 acteristics of energy model for developing countries 3. Back casting: are also extracted from the related literature in sec- This approach is used to determine the conditions tion 3. of a desired future and to define steps to attain a desired vision of the future. This is an alternative to 2. Approaches of characterizing energy traditional forecast which relies on what is known models today and the future is viewed as a continuum of The energy models existing in the literature can be past or present. Back-casting is a planning method- categorized via different ways; however, these cate- ology under uncertain circumstances that is particu- gorizations are related to each other. In the follow- larly helpful when problems are complex, and there ing, we will discuss each of these ways in more is a need for major change and in cases in which it details. is risky to view the future just in the mirror of the past (Holmberg, 2000). Model type Descriptive or prognostic models depict or describe Modelling paradigm how things actually work or might work, and The difference between top-down and bottom-up answer the question, ‘What is this?’ In comparison, models is related to the technological and sectoral normative models are prescriptive and suggest what aggregation. A broader economy is investigated by should be done (how things ought to work) accord- use of top-down models in order to examine effects ing to an assumption or standard. The first between different sectors and they do not consider approach belongs to the concept of planning as details of energy production technologies. Smooth reaction and the second approach involves goal production functions are used to represent energy attainment and assumes autonomous planning or sectors in an integrated way. In such models, sub- planning as an action. Descriptive models comprise stitution is determined by elasticity. On the other different methods of econometrics or simulation, hand, technologies are represented in detail in bot- while normative models lie within the scope of opti- tom-up models, but they miss to take into account mization. economy-wide interactions such as price distortions (Bohringer, 1998). Purpose Grubb et al. (1993) stated that the top-down Energy models are usually designed to address spe- approach addresses the ‘descriptive’ economic par- cific questions and hence, are only suitable for the adigm, while the bottom-up approach is associated purpose they were developed. For our categoriza- with– but not exclusively restricted to– the ‘norma- tion, we will make a distinction between three pur- tive’ engineering paradigm. In the economic para- poses i.e., Prediction/Forecasting, Exploring, and digm, technology is considered as a set of process- Back-casting purpose as follows (Beek, 1999): es by which inputs such as capital, labour, and ener- gy can be transferred into useful outputs and the 1. Prediction/forecasting ‘best’ or most optimal techniques are defined by Basically, forecasting focuses on extrapolation of efficient markets. However, in the engineering par- trends found in historical data. A prior condition for adigm, the developed model is independent of this method is that the critical underlying parame- observed market behaviour. In other words, the ters remain constant. Therefore, this method can be economic paradigm is based on market behaviour applied for analysing relatively medium to long and aggregated data, while the engineering para- term impacts of actions. In fact, forecasting is about digm tends to ignore existing market constraints and what things will look like in the future and the uses disaggregated data. For example, in the top- method used for forecasting depends on the situa- down approach the key question is ‘By how much tion. Prediction uses past observations to extrapo- does a given energy price movement change ener- late future short-term observations. Long-term fore- gy demand or energy-related carbon emission?’ In casting is usually made by econometrics methods contrast, in bottom-up the question is ‘How can a and short-term prediction uses extrapolation meth- given emission reduction task be accomplished at ods (Armstrong, 2001). minimum costs?’(e.g., see Frei et al. 2003). While the traditional top-down approach follows 2. Exploring: an aggregated view and believes in the influence of Scenario analysis is utilized for exploring the future. price and markets, the bottom-up models focus on In this method, a limited number of developed sce- the technical characteristics of the energy sector. narios are compared with a Business As Usual Hybrid models try to bridge the gap between top- (BAU) reference scenario. In developing scenarios, down and bottom-up by including elements of both some assumptions like economic growth and tech- approaches. They attempt to combine the benefits nological progress, which are not relied on the of both top-down and bottom-up modelling parameters extracted from the past behaviour, are schemes using each modelling vision where appro- made. priate and a modular structure to integrate the dis- 102 Journal of Energy in Southern Africa • Vol 25 No 4 • November 2014 parate systems. The hybrid models undertake to in finding a good choice out of a set of alterna- consider economic effects on model outcomes tives by minimizing or maximizing one or some through taking into consideration the market real functions. Input values are selected from an behaviour. Market behaviour is a result of interac- allowed set and must satisfy some constraints. tions among the economy sector, supply