Location optimization of solar plants by an integrated hierarchical DEA PCA approach
Introduction
Newly emerging renewable alternative energy resources are expected to take increasing role in the energy scenarios of the future energy consumptions and specially environmental and technical benefits of solar energy have made it as a promising alternative for conventional energy resources. Solar energy is the most ancient source, and it is the root material for almost all fossil and renewable types. Solar energy is freely available and could be easily harnessed to reduce our reliance on hydrocarbon-based energy by both, passive and active designs. Utilization of solar energy is going to develop in allover the world and one of the most used applications of solar energy is in the solar plants. These facilities have a great potential for supplying energy in the remote and shiny regions. Determining the priority of different locations has special importance for utilizing solar systems. To this end, we have considered location selection of the solar plants in this article.
Location decisions are used in any field of facility establishment. The term “location” refers to the modeling, formulation, and solution of a class of problems that can best be described as sitting facilities in some given space. The bibliography by Domschke and Drexl (1985) lists more than 1500 references dealing with location and layout problems, and many more contributions have appeared since then. There are 4 components that characterize location problems; these are: (1) customers, who are presumed to be already located at points or on routes, (2) facilities that will be located, (3) a space in which customers and facilities are located, and (4) a metric that indicates distances or times between customers and facilities (Bhatnagar and Sohal, 2005). Applications of location problems abound: they range from gas stations and fast food outlets to landfills and power plants.
A survey of many distinct applications of location models is provided by Eiselt (1992), ranging from traditional applications involving newspaper transfer points (Jacobsen and Madsen, 1980), solid waste transfer points (Marks and Liebman, 1971; Wirasinghe and Waters, 1983), bank branches (Hopmans, 1986), and motels (Kimes and Fitzsimmons, 1990) to the more unusual location problems such as the location of a church camp (Huxley, 1982), the determination of apparel sizes (Tryfos, 1986), ingot sizes (Vasko et al., 1987), and the location of rain gauges (Hogan, 1990). For a list of new location applications, readers are referred to Current et al. (2002).
Whereas routing problems are typically found on the lower end of the strategic–tactical–operational continuum (meaning that they are typically reasonably well defined and measurable), location problems are likely to have multiple objectives; fuzzy and ill-posed, making it much harder for the analyst to model them. One approach that appears to be promising is the use of tools from decision analysis; see, e.g., Rey et al. (1995), who use the ELECTRE1 outranking method to arrive at a solution that reconciles the many different criteria included in the problem. Also the use of multicriteria methods such as MCDM2 and MADM3 have increased for considering different parameters for location optimization of the plants. In this article we have used data envelopment analysis (DEA) as a multicriteria method for location of solar plants. In the following different factors and criteria that are used for location of the plants have been mentioned.
From the viewpoint of incorporating qualitative factors in the location decision, the most widely used technique is a weighted checklist approach in which various important but diverse factors like proximity to customers, business climate, legislation, tax incentives and other support factors are rated on a weighted scale and combined into an aggregate score. The selected site is the one with the best aggregate score. Details and applications in a wide variety of industries are reported by Bowersox and Closs (1996), Chase et al. (1998), Ballou (1999), and Krajewski and Ritzman (1999) among others. Such an approach can lead to subjective results depending on the preferences of the decision maker. Moreover, there has been small attempt in previous research to measure the effectiveness of such a weighting mechanism.
Schmenner (1982) tested the significance of qualitative variables for the plant location decision and reported a comprehensive survey of the plant location/re-location practices among Fortune 500 companies in the US. The study identified favorable labor market, nearness to market, quality of life in the area, nearness to supplies, and low labor rates as the most important variables considered by managers in the location decision. MacCormack et al. (1994) examined the impact of the global trading environment, new production systems, and new technologies on the plant location decision. The authors suggested that existing literature approached the plant location problem narrowly, focused on quantitative data such as transport costs, exchange rates, taxes, and labor rates, as opposed to qualitative factors such as infrastructure, worker skills, local government regulations, and access to suppliers.
In this article we have proposed some of the most important factors that are effective for selecting the location of solar plants. These parameters are then used in a hybrid methodology for indicating the priority of nominal regions for location of a solar plant. Our prescribed type of solar plant which has been considered for this study is a kind of concentrator plant with trough mirror. This type of plant works through circulating synthetic oil in an integrated network and generating hot steam by using a heat exchanger.
Section snippets
Methodology
A set of technical, geographical, and social factors for location optimization of solar plants are considered in this paper. Furthermore, a new methodology based on DEA for assessment of the solar plant establishments in the selected regions is presented for location optimization of solar plants with regard to characteristics of each area. Based on this methodology, we have defined the concept of location as efficiency of a region for location of solar plants. DEA has been used as an
The case study
We have applied the integrated hierarchical DEA model for location optimization of solar plants in 25 nominal cities in Iran. We have considered 6 different regions in each city. Table 2 presents the classification of the 6 regions in each city. This was accomplished by the aid of experts in the field. There are 150 locations in Iran for solar location problem. This shows the wide coverage of the model.
Verification and validation of the results of model
The DEA results are verified and validated by 2 multivariable ranking methods namely principal component analysis (PCA) and numerical taxonomy (NT). Furthermore, the results of level 2 DEA for determining the priority of the cities for construction of solar plants are compared with the results of PCA and NT. The structure of PCA and NT for determining the priority of 25 nominal cities has been presented in the next sections.
Analyzing the results of models
For analyzing the acquired results, critical parameters for each DMUs in level 1 are identified. Population and human labor is the most critical indicator for about 76% of the cases (114 regions). In addition, distance of power distribution networks is the second most critical indicator for about 20% of the cases (30 regions). This shows the importance of human factors such as population and human labor for selecting the location of solar plants within a city in Iran.
For the results of level 2,
Conclusion
Environmentally friendly benefits of solar power plants make them very desirable as an alternative source of energy. Hence, determination of the optimum locations for use of this resource is a vital decision. Generally, solar radiation as a primary tool is used for determining the optimum locations for power plants. Therefore in this approach some local and social considerations are ignored. Some criteria such as population of the region, geological and geographical considerations, and involved
References (53)
- et al.
An ANN approach for predicting global radiation in locations with no direct measurement instrumentation
Renewable Energy
(1998) - et al.
Solar radiation for Dhahran, Saudi Arabia
Energy
(1986) - et al.
Estimation of daily and monthly direct, diffuse and global solar radiation from sunshine duration measurements
Solar Energy
(1984) - et al.
Supply chain competitiveness: measuring the impact of location factors, uncertainty and manufacturing practices
Technovation
(2005) - et al.
Measuring efficiency of decision making units
European Journal of Operational Research
(1978) - et al.
An activity analysis model of the environmental performance of firms—application to fossil-fuel-fired electric utilities
Ecological Economics
(1996) A spatial interaction model for branch bank accounts
European Journal of Operational Research
(1986)- et al.
Estimation of hourly global photosynthetically active radiation using artificial neural network models
Agricultural and Forest Meteorology
(2001) - et al.
Estimation of global solar radiation using artificial neural networks
Renewable Energy
(1998) - et al.
Evaluation of travel demand measures and programs: a data envelopment analysis approach
Transportation Research A
(1998)