یک مدل مبتنی بر برنامه ریزی پویا تصادفی برای تعیین سیاست بهینه ذخایر استراتژیک نفت چین
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|25498||2009||10 صفحه PDF||سفارش دهید||6471 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Energy Policy, Volume 37, Issue 11, November 2009, Pages 4397–4406
China's Strategic Petroleum Reserve (SPR) is currently being prepared. But how large the optimal stockpile size for China should be, what the best acquisition strategies are, how to release the reserve if a disruption occurs, and other related issues still need to be studied in detail. In this paper, we develop a stochastic dynamic programming model based on a total potential cost function of establishing SPRs to evaluate the optimal SPR policy for China. Using this model, empirical results are presented for the optimal size of China's SPR and the best acquisition and drawdown strategies for a few specific cases. The results show that with comprehensive consideration, the optimal SPR size for China is around 320 million barrels. This size is equivalent to about 90 days of net oil import amount in 2006 and should be reached in the year 2017, three years earlier than the national goal, which implies that the need for China to fill the SPR is probably more pressing; the best stockpile release action in a disruption is related to the disruption levels and expected continuation probabilities. The information provided by the results will be useful for decision makers.
China's demand for oil has skyrocketed with the rapid development of its economy. In 2002, China's oil consumption surpassed that of Japan, and became the second largest oil consuming country in the world, just behind the US. Simultaneously, its dependency on foreign oil is rapidly increasing. In 1993, China became a net importer of oil, and in 2007 its dependency on foreign oil increased to 46.05%. Furthermore, the oil that China imports comes from highly centralized sources, with nearly 40% coming from the Middle East. This high dependency on oil from overseas and centralization of oil imports poses risks for China's oil supply. If oil supply is interrupted, it will be detrimental to China's economic security and social stability. It is thus critical to take measures to ensure the security of China's oil supply. Strategic Petroleum Reserve (SPR) is a basic strategy for mitigating the effects of a disruption to the oil supply and it has proved to be rather effective in many developed countries. The National People's Congress of China named the development of a SPR a primary goal in 2001 and the Chinese government has been preparing for the establishment of its SPR since March 2004. According to the remarks made by Chen Deming, the former vice minister of the National Development and Reform Commission (NDRC) in November 2007, the planned goal is to build a SPR equivalent to 30 days of net oil import by 2010, and to store sufficient amount to cover about 90 days of net imported oil by 2020 (China News Service, November 23, 2007), which is about 320 million barrels according to China's oil import level in 2006 (State Statistical Bureau, 2007). The first SPR base of China, located in Zhenhai, was finished in August 2006. This base is one of the four stockpile bases included in China's first SPR project and has a capacity of 5.2 million cubic meters. That China should establish a SPR is indisputable, but some questions remain regarding the cost of stockpiling plans, such as how large the SPR should be, whether the government's goal is the optimal, when the best time for China to fill the SPR capacity is, and what the optimal acquisition and drawdown strategies are. Hence, it is vital to devise an optimal SPR policy for China, including the optimal size and best buildup and drawdown rates, to minimize the cost or to maximize the benefit of establishing its SPR. Many countries, including the US, suffered greatly from the oil crisis in the 1970s, triggering a wave of studies concerning SPR and energy security. Balas (1981) studied a short-term interaction between oil importing nations and a politically motivated cartel that takes advantage of disruptions to inflict economic losses on importing nations. In this study, Balas investigated the ‘deterrence effect’ of a SPR and found that a stockpile not only reduces the economic losses from a disruption, but also lessens the likelihood of a disruption. Teisberg (1981) developed a dynamic programming model for the SPR of the US, which could be used to determine optimal acquisition and drawdown strategies for two different states of the world oil market: normal and disrupted. Samouilidis and Berahas (1982) established a decision tree model to evaluate different scenarios for a SPR based on a cost function that includes procurement, maintenance, and shortage costs. Chao and Mane (1983) developed a multi-period dynamic programming model for obtaining the optimal stockpiling and petroleum usage rates based on their analysis of the oil supply policies of the US. Hogan (1983) extended Teisberg's model of US stockpiling to a Stackelberg model to examine the interactions between two consuming countries, where one follows the other's lead. Samouilidis and Magirou (1985) presented a concise analysis for the optimal selection of the size of a SPR for a small country based on the work of Samouilidis and Berahas (1982). Oren and Wan (1986) presented a non-linear programming model for conducting a steady-state analysis on the optimal size, fill-up, and drawdown rates for a SPR under variety of different supply and demand conditions. Murphy et al. (1987) presented a Nash dynamic game model of interactions between oil inventory and tariff policies for oil-importing countries to analyze their SPR policies. Zweifel and Bonomo (1995) established a model that considers multiple risks to energy supplies to illustrate that one-dimensional rules such as an “oil reserve for 90 days” turn out not only to be suboptimal but also suggest adjustments that make them even more suboptimal. All these above-mentioned studies provide valuable information for China's SPR research. The majority of studies concerning China's SPR only contain qualitative analysis. However, more quantitative studies have appeared in recent years due to greater concerns about the oil supply security of China. Wei et al. (2008) conducted an empirical analysis of the optimal SPR size for China based on a decision tree model. Wu et al. (2008) presented an uncertain programming model for analyzing acquisition strategies for China's SPR. However, these studies considered the optimal size or acquisition strategies alone, and did not present a comprehensive SPR policy. Referring to the above-mentioned models, we developed a stochastic dynamic programming model for China, which we termed Strategic Oil Stockpiling for China (SOSC) to determine the optimal SPR size and strategies that include both the acquisition and release strategies for several different situations for the time horizon from 2009 to 2039. We believe that the establishment of China's SPR should be principally based on the normal domestic supply and normal import. This means that in order to avoid a potential rapid increase of oil prices due to China's oil stockpiling action, China should not increase the amount of oil it imports to stock its SPR. Our model is based on this concept. In the following sections we develop our SOSC model, which is based on stochastic dynamic programming. In Section 2, we develop our SOSC model step by step. The implementation of this model is described in Section 3. Then, Section 4 presents illustrative results obtained by implementing our model. It also discusses the implications of the results. Finally, in the last section we give some conclusions and make suggestions for policy-makers based on the results of our model.
نتیجه گیری انگلیسی
Determining the optimal SPR policy is a very important problem, but it is also very complicated. Many factors need to be considered to evaluate the optimal SPR policy. Consequently, we have made many assumptions, which might omit many features of China's stockpile problem. But our application of DP to this problem considers the most critical factors that significantly affect the optimal stockpile policy of China, and it generates many reasonable outputs for the optimal action with respect to each time period, each market state, and each existing stockpile size. We believe that it can provide decision makers with a useful reference. The results obtained in our study suggest the following conclusions about China's SPR policy: • If business develops as usual, China should build up its SPR with an initial acquisition rate of around 100 million barrels in year 2009. In the following years, the buildup rate should be lower and the optimal SPR level of about 320 million should be reached in the year 2017, which is three years earlier than the government's goal. • If the disruption probabilities are high, the optimal SPR size for China is up to 550 million barrels in the case we presented. In this case, the optimal choice for the first year's acquisition is 130 million barrels. • If a slight disruption occurs, the amount to release should vary with the year when the disruption takes place. Our empirical results show that when a slight disruption occurs nearer to the terminal time, China should release more SPRs. More comprehensively, the optimal release decision when a disruption occurs should differ with different disruption levels and expected continuation probabilities. If the disruption is slight, the policy makers of China should release less existing stockpiles for high continuation probabilities case than for low probabilities case, since holding more stockpiles to the next period becomes much more important when the disruption is expected to be longer. If the disruption is serious, this effect exists as well. China needs to release all the stock in the base case but to conserve some SPRs in the high continuation probability case. The comparison of the base case with the case of changed time horizons implies that a shorter time horizon will perhaps reduce the need for SPR establishment. So it may be wise for China to try to remove the dependency upon oil as soon as possible through feasible actions, such as the wide usage of new energy technologies, the significant improvement of its energy efficiency, and so on. To ensure the energy supply security of China, the establishment of a SPR is an important strategy. Because of the high potential cost of establishing strategic oil stockpiles, it is necessary to devise an optimal strategy for determining the optimal SPR size. Our SOSC model develops a stochastic DP problem to generate China's optimal SPR policy, and discusses appropriate buildup or drawdown strategies for several cases. However, this model did not consider stockpiling by private companies, which may affect the government's optimal decision. This is an area we intend to investigate in the future.