ترکیب رفتار خرید فن آوری در مدل های انرژی سهام داخلی بلند مدت بر اساس UK برای ارائه تجزیه و تحلیل سیاست های بهبود یافته

The UK has a target for an 80% reduction in CO2 emissions by 2050 from a 1990 base. Domestic energy use accounts for around 30% of total emissions. This paper presents a comprehensive review of existing models and modelling techniques and indicates how they might be improved by considering individual buying behaviour. Macro (top-down) and micro (bottom-up) models have been reviewed and analysed. It is found that bottom-up models can project technology diffusion due to their higher resolution. The weakness of existing bottom-up models at capturing individual green technology buying behaviour has been identified. Consequently, Markov chains, neural networks and agent-based modelling are proposed as possible methods to incorporate buying behaviour within a domestic energy forecast model. Among the three methods, agent-based models are found to be the most promising, although a successful agent approach requires large amounts of input data. A prototype agent-based model has been developed and tested, which demonstrates the feasibility of an agent approach. This model shows that an agent-based approach is promising as a means to predict the effectiveness of various policy measures.

Energy efficiency first came on to the political agenda in the 1970s as a response to the oil crises. Since then it has been gradually gaining in importance. Today, the two main concerns are energy security – ensuring there will be continuous and sufficient supplies of energy; and climate change – concerns over emissions from energy generation (DECC, 2011a). In the UK, the main focus regarding emissions is on CO2 and in the 2008 Climate Change Act the UK Government has committed the country to an 80% reduction target by 2050 from a 1990 base level. Approximately 28% of energy use is in the home (DECC, 2011b). This can be further broken down to some 56% for space heating, 26% hot water, 15% lighting and appliances and 3% for cooking (DECC, 2011c). Therefore, if an 80% overall target is to be met, significant reductions will be required in the domestic sector. Modelling can be used to help in planning a suitable pathway to 2050 in order to meet the carbon reduction target; for instance, by considering the impact of projected population changes, or to predict the effectiveness of different policy measures. There are two broad types of models: top-down models that are macro-economics based and typically operate on a whole economy basis; and bottom-up models operating at the micro-level and usually sector specific, e.g. domestic dwellings, transport, industry, etc. This paper therefore provides a comprehensive review of existing models that include domestic dwellings, and their various purposes, together with a discussion of the respective strengths and weaknesses of their different methods. To conclude, recommendations are made for new techniques that could be used to improve on existing methods.

This research has shown that bottom-up modelling is the most appropriate method for predicting dwelling stock emissions to 2050. However, approximately two-thirds of homes belong to owner-occupiers and energy efficiency improvements will only happen when those individuals decide to invest in green technologies. Therefore models need to be able to incorporate heterogeneous individual buying decisions. Agent-based modelling has been identified as the most promising method to include such micro-level behaviour, a feasibility study including a prototype model indicated that this will be possible. The challenge for future research will be to understand the individual’s decision making process. There should be two parts to this: firstly, to determine the trigger points that cause a decision to be made; secondly to identify the different factors affecting the decision making process and the weighting that should be applied to each factor. A number of technologies still have very small installed numbers – less than 1% of the housing stock – in those cases the installations will typically have been from environmentally aware early adopters or from new builds where the owner-occupier will have had no input in the technology buying process. Therefore it will be difficult to predict the responses of the bulk of the population based on the actions of the early adopters. This suggests that, initially at least, a substantial amount of data will need to be collected via simulated buying experiments, with the obvious caveat that stated preferences will not be identical to real world decisions. This therefore further suggests the need for longitudinal studies and calibration against real world data as it becomes available. Therefore, if suitable data can be collated, agent-based modelling has a lot of promise for the analysis of pathways to 2050 and considering the cost effectiveness of both individual policies and packages of policy measures. Scenarios can be constructed with different sets of policies, e.g. subsidies, taxation, grants, loans, etc., and the diffusion rates of different technologies can then be tracked over the lifetime of the model to find cost effective pathways to 2050. Therefore a full agent-based model, with a comprehensive dwelling stock and suitable decision making data, will be usable by policy makers to simulate the effectiveness of different sets of policy interventions; this will enable policy makers to test current policies and compare them with alternative options in an effort to maximise emissions reductions, whilst simultaneously minimising the associated costs. The detailed information of how to operate this model and comprehensive scenario-based case studies relating to the UK energy policies will be presented in a forthcoming paper.