Abstract
An important decision for policy makers is selecting strategic petroleum reserve sites. However, policy makers may not choose the most suitable and efficient locations for strategic petroleum reserve (SPR) due to the complexity in the choice of sites. This paper proposes a multi-objective programming model to determine the optimal locations for China’s SPR storage sites. This model considers not only the minimum response time but also the minimum transportation cost based on a series of reasonable assumptions and constraint conditions. The factors influencing SPR sites are identified to determine potential demand points and candidate storage sites. Estimation and suggestions are made for the selection of China’s future SPR storage sites based on the results of this model. When the number of petroleum storage sites is less than or equals 25 and the maximum capacity of storage sites is restricted to 10 million tonnes, the model’s result best fit for the current layout scheme selected thirteen storage sites in four scenarios. Considering the current status of SPR in China, Tianjin, Qingdao, Dalian, Daqing and Zhanjiang, Chengdu, Xi’an, and Yueyang are suggested to be the candidate locations for the third phase of the construction plan. The locations of petroleum storage sites suggested in this work could be used as a reference for decision makers.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Oil, known as the lifeblood of industry, is a strategic raw material as well as a major source of energy supporting economic and social development (Karl 2007). The first of three oil crises began in 1970s, causing higher prices and a stagnant global economy (Helm 2002). Accordingly, western countries began to store emergency petroleum reserves (Fan and Zhang 2010). With the expansion in scale, a strategic petroleum reserve is considered as an effective tool to improve energy security and alleviate price fluctuations (Hubbard and Weiner 1985). According to the reserve agreement, 28 countries of the International Energy Agency were required to hold 90 days of net oil imports for their respective countries. As a second-largest oil consumer in the world, energy security in China is among the most serious challenges due to its increasing oil dependency (Dong et al. 2017). Additionally, the geopolitical impacts of OPEC on oil supply and prices are a significant factor for China to have its own SPR (Chen et al. 2016). Following other nations’ establishment, China has built its own SPR to ensure that its oil supply will not be disrupted. The preliminary work on China’s SPR was prepared in 1993, approved by the government in 2003, and commenced in 2004 (Jiao et al. 2014). The timeline of China’s SPR program is shown in Fig. 1. According to China’s stated policy on SPR, the country is expected to build storage capacity equivalent to 90 days of its net imports in three phases over 15 years. In reality, China has delayed completion of the second phase until 2020 (Park 2015). Generally, the establishment of an SPR in China has helped in dealing with oil crises and ensuring energy and economic security.
An SPR project is a complex system, and many practical questions need to be answered (Davis 1981). The selection of the petroleum reserve sites has a profound effect on SPR program operation. Not all locations are appropriate, so it is important for policy makers to identify the most suitable and efficient locations. Based on China’s SPR construction status (Fig. 2), four reserve sites were selected in Phase I, locating in the coastal cities of Zhenhai, Zhoushan, Huangdao, and Dalian. Eight reserve sites in Phase II are planned for inland areas, including Tianjin, Jinzhou, Dushanzi, Xishan, Lanzhou, Jintan, Huizhou, and Zhanjiang. The third selection will likely be located in Wanzhou, Henan, Caofeidian, and Tianjian (Wu and Wang 2012). Three criteria affect the location selection at which China stores its purchased oil. The first is transportation convenience. Most sites are not only key petroleum import harbors but also near substantial demand centers, so the oil can be transported with low transportation costs and a short response time; the second factor is achieving maximum safety. The apparent move from coastal cities to inland is in consideration of increased safety for the SPR in underground tanks scattered around the country (Wu 2014). Third, the location distribution is closely associated with the layout of oil consumption and the routes of imports (Zhu 2007). Today, the preparatory work for the third phase has been launched and the sites selections are still being determined. How to decide on the storage sites rationally and scientifically is a practical issue we need to consider to achieve the full strategic, economic, and social benefits of the SPR program.
SPR site location is categorized as an emergency facility location problem (Farahani et al. 2012). The p-median method, p-center method, and covering method are widely used in the literature to decide emergency facility locations (Toregas et al. 1971; Roth 1969; Li et al. 2011). However, these discrete methods are proposed based on single objective such as minimum time and distance. With regard to the SPR site selection, there is limited related research. Zhang et al. (2008) presented an uncertain planning model to determine the location of SPR sites with minimal cost of transport based on hybrid intelligent algorithms. Based on fuzzy comprehensive evaluation and maximal covering models, Chen suggested 14 candidate cities for China’s second and third SPR construction (Chen 2010). Compared with these methods, the prominent advantage of the multi-objective model is that it takes full consideration of cost and time (Wilson et al. 2013). The comparisons between this study and other facilities location study are given in Table 1 (De Vos and Rientjes 2008). Accordingly, this paper presents a time and cost multi-objective programming model to identify the optimal SPR storage site locations and guide the construction of SPR in China.
2 Factors influencing SPR storage site location
Generally, the location decision for an SPR storage site should consider the economic considerations and other non-economic considerations, such as safety, efficiency, and fairness. Among them, cost and time minimization is the two most important factors in the choice of a particular place to locate the SPR. In other words, the SPR site should be optimally located to achieve the minimum response time and construction cost (Dong et al. 2013). Because the SPR is characterized by providing emergency oil, the time to access the emergency oil supply must be as short as possible (Oregon 2003). Furthermore, as a construction project, reducing the construction cost is regarded as a basic goal in SPR site selection. With regard to the factors affecting SPR storage sites, it is necessary to take into consideration all the relevant factors. Niu indicated that five factors affected the location decision, including military benefit, economic benefit, transportation convenience, sustainability, and compatibility (Niu et al. 2010). According to Li and Tan, SPR storage site location should satisfy two principles: absorbing the imported oil and quickly transporting it to a refining center (Li and Tan 2002). In addition, the accessibility of the oil resource, the convenience of transportation, the superiority of storage, and the effectiveness of releasing the emergency oil should be taken into account in location decisions. Based on the previous work, the major factors influencing location are discussed below:
-
Availability of petroleum resources in determining the reserve location for the SPR, the availability of petroleum resources is of vital importance, because the availability can ensure emergency supply and reduce the cost of production. The basic requirement for a reserve site is that it can hold enough petroleum to address supply disruptions (Bai et al. 2016a, b).
-
Proximity to refining centers policy makers must consider nearness to refining centers. Petroleum is difficult to transport over long distances so an SPR should be located in close proximity to refining centers. Locating reserve sites near refining centers can reduce transportation cost and reduce response time (Majid et al. 2016).
-
Convenience of transportation convenience of transportation also influences the SPR storage sites. The four modes of petroleum transportation (pipeline, water, rail, and road) play a significant role. Thus, the junction points of these transport types become priority areas for SPR location (Niu et al. 2013).
-
Petroleum storage consideration petroleum storage consideration refers to storing the petroleum safely in natural and climatic conditions. These factors can influence the location of an SPR storage site. Stable geological conditions with underground petroleum storage caverns can provide an added advantage over conventional storage tanks in ensuring the storage security of the SPR (Tillerson 1979).
-
Distance between reserve site and demand center the distance between the reserve site and demand center is associated with the effectiveness of SPR release. A long distance not only causes response time delay but also increases transportation cost. Accordingly, the reserve site should be located as close as possible to the demand center (Williams 2008).
-
Strategic considerations as petroleum is a strategic commodity, strategic consideration is important in determining the SPR storage site location. The location decision should be consistent with the strategic planning for development with coverage as complete as possible (Bai et al. 2016a, b).
As mentioned above, the influencing factors address three aspects: demand point information, reserve site information, and information related to both (Liu et al. 2014). The information about demand points refers to the geographic location and the amount of emergency petroleum demand. The reserve site information contains the location and storage capacity. The distance between demand point and reserve site and the transportation conditions also have an effect on SPR site selection.
3 Methodology
3.1 Data
3.1.1 Demand points information
Based on the analysis of factors affecting site location determination and following the principle of proximity, we collected the information about SPR emergency demand points in terms of geological location and demand amount. According to the Chinese Statistical Bureau report in 2014, there were over 260 refineries (CNPC 2015). As the largest refinery companies, CNPC and Sinopec accounted for 28% and 38%, respectively, of the refining capacity. Furthermore, CNOOC, Yanchang Shaanxi Petroleum, and local enterprises played a significant role in Chinese refining (The Oxford Institute for Energy Studies 2016). Therefore, we selected 64 refineries located in 52 cities as the SPR demand points, as shown in Table 2. The demand amount is calculated by indicators based on the petroleum self-sufficiency ratio, risk probability, and transportation distance (Martínez-Palou et al. 2011). To ensure the 90 days of SPR supply, the reserve amount is assumed to equal the demand amount. Due to the limitation of data availability, this demand amount for each point is calculated based on data from 2012 when 270 million tonnes of imported crude oil were imported. Furthermore, the distribution of emergency demand of each city should follow the industry production ratio. The results are given in Table 3.
3.1.2 Candidate reserve site location
According to the regulation of National Petroleum Reserve in China, the reserve location should ensure the supply of emergency SPR efficiently and safely. Based on Liu, the candidate reserve site location is determined by three indicators: the flow function, the spatial structure, and the flow track. Table 4 shows the function types of crude oil flow in China based on calculation of the oil self-sufficiency ratio and liquidity ratio. The spatial structure of oil flow is consistent with the petroleum distribution in China, including source system, transit system, and sink system, as shown in Fig. 3. Taking the above factors into consideration, 55 prefecture-level cities are listed as candidate SPR sites in China. The detailed information is shown in Table 5. Specifically, the candidate sites include 17 coastal harbors, 18 input origins, 5 terminal hinges, 6 inland ports, and 9 intersecting regions. The selected site locations cover nearly all the key areas of petroleum resource mobility.
3.1.3 Distance calculation
Petroleum transport is a major component of an SPR system with a range of transportation options available, including railway, pipeline, highway, and water networks. Considering the wide range of cities involved and the incomplete information on pipeline networks, we calculated the shortest distance between the demand point and candidate site based on the shortest path theory of railway mileage. The result is shown in Table 6.
3.1.4 Selection criterion
In the SPR site selection model, four important parameters should be discussed. The first parameter refers to the time the reserve site needs to respond to an emergency demand. In terms of economic and safety considerations, the candidate location can cover the demand point once. The storage capacity parameter of candidate reserve sites depends on the maximum and minimum capacity. Considering the limitation of SPR storage capacity, two maximum storage scenarios of 10 and 8 million tonnes are discussed based on Liu, which should outweigh the minimum capability (≥1 million tonnes). Then, the emergency response time should be evaluated on the basis of the average railway speed (90–100 km/h), so the response range is assumed to 900 km. The last parameter is the reserve numbers. Too few or too many reserve sites will affect the scheme. This paper will discuss 20 or 25 reserve sites considering the current status of and prospects for SPR construction.
3.2 SPR site location model
3.2.1 Problem description
First, we define two sets, G and F, which are the function of potential demand points and reserve sites, respectively. G and F assume that the geographic location of the demand points and reserve sites, the distance between them, and the transportation costs are known. To provide emergency petroleum with minimal time and cost, the problem is to determine numbers of reserve sites from among the optional locations.
3.2.2 Assumptions
We have based our model on several assumptions as follows:
-
The model is categorized to the discrete allocation problem, comprising various optional demand points and reserve sites.
-
The demand points are evaluated by geographic location and amount of petroleum demand.
-
Petroleum demand volume for each point is fixed.
-
The storage capacity for each optional reserve site is limited by the maximum and minimum volume. The construction scale of reserve sites is restricted to the storage capacity.
-
A demand point is supported by the reserve site under the assumption that the coverage frequency equals 1.
-
The amount of petroleum supply from the reserve site should meet the demands under the coverage.
-
The supply volume of the selected reserve site should be less than or equal to its maximum storage capacity.
-
The transfer velocity is constant; that is, the transfer distance has an equivalence relation with transfer time.
-
In reality, the location and construction of reserve sites satisfy the standard of technology that all the reserve sites are consistent with the technical standards.
3.2.3 Symbol
The SPR storage site selection model can be expressed as follows:
-
(1)
Indices symbols
-
\(G = \left\{ {G_{i} \left| {i = } \right.1,2, \ldots ,m} \right\}\) is a set of SPR demand points;
-
\(F = \left\{ {F_{j} \left| {j = } \right.1,2, \ldots ,n} \right\}\) is a set of optional SPR storage sites;
-
-
(2)
Parameters symbols
-
\(w_{i}\) is crude demand of SPR \(i \in G\);
-
\(\mathop C\nolimits_{j}^{ - }\) and \(\mathop C\nolimits_{j}^{ + }\) are the minimum and maximum storage capacity of reserve sites \(j \in F\);
-
\(P\) is the constant parameters of the optional reserve sites;
-
\(d_{ij}\) is the distance between demand point \(i\) and candidate reserve site \(j\);
-
\(\lambda\) is the emergency response time;
-
\(C\) is the total volume of SPR storage capacity in the country;
-
-
(3)
Variables symbols
-
$$x_{ij} = \left\{ {\begin{array}{*{20}l} 1 \hfill & {{\text{if the demand point }}i{\text{ is assigned to reserve site}};} \hfill \\ 0 \hfill & {\text{otherwise}} \hfill \\ \end{array} } \right.$$
-
$$y_{ij} = \left\{ {\begin{array}{*{20}l} 1 \hfill & {{\text{if the reserve site }}j{\text{ is selected}};} \hfill \\ 0 \hfill & {\text{otherwise;}} \hfill \\ \end{array} } \right.$$
-
\(L\) is the maximum distance between demand point \(i\) and its covered reserve site \(j.\)
-
3.2.4 Objectives
An SPR storage site location model is a decision-making tool for identifying reserve locations in a landscape to achieve two objectives: minimum response time and minimum construction cost. Based on the current construction situation and the experience of other countries, the location should be near convenient traffic connections and refineries, which can decrease the response time and save costs. The first objective is the shortest response time, which means the distance between the emergency demand point and its nearest emergency reserve point is minimized. The formula can be described by the p-center model as follows (Current et al. 1990):
The second objective is minimum response cost, which aims to ensure that the total weighted distance between each demand node and its closest reserve site is minimized. The p-median model is applied to minimize the response cost, which can be expressed as follows (Mladenović et al. 2007):
3.2.5 Constraints
A modeling constraint is a requirement for a candidate solution based on the objective function. In the SPR site selection model, the candidate base should be subject to the following constraints. Constraint (3) indicates that the distance between the SPR site and SPR demand should be less than the maximum L; Constraint (4) ensures the existing relation until the candidate site is selected. Constraint (5) means the frequency of the SPR site will be covered. Constraint (6) limits the total number of selected SPR sites. Constraints (7) and (8) present the volume limitation for SPR demand and supply, respectively.
3.2.6 Multi-objective model for SPR site selection
Combining the objectives and constraints, a time–cost multi-objective model for SPR site selection is elaborated. Furthermore, some constraints should be clarified in considering the current construction situation. Specifically, Constraint 4 should be modified so the value is less than P, and Constraint 6 indicates the volume of SPR demand, which is equal to C.
3.2.7 Algorithms
As a discrete objective decision-making model, the algorithm of goal programming is presented to solve the multi-objective problem (Aouni and Kettani 2001). Then, the analytic hierarchy process is applied to determine the weights of each objective. Finally, Lingo software is used to obtain the results.
4 Results and discussion
4.1 Results
Based on the SPR site location model, the optimal solution for the first and second objective function is calculated by Lingo, \(V_{1}^{*} = 812\) and \(V_{2}^{*} = 999,194.1\). Under the conditions of the emergency delivery distance of 900 km, the total number of reserve sites (20 or 30), and the storage capacity (800 or 1000), Figs. 4 and 5 present the results of optimal SPR site locations selected from candidate cities under two scenarios. Considering the equal significance of fairness and efficiency, the weight value of the two objectives equals to 0.5 applied in the multi-objective programming model, which consults Zhang et al. for a reference (Zhang et al. 2008).
4.2 Discussion
4.2.1 Four-scenario discussion
Table 7 shows the selected results under four scenarios. For scenario 1, the assumption is that the upper limits of the reserve numbers and storage capacity are 20 and 1000 (\(P \le 20,\;C \le 1000\)), respectively. Nineteen cities are selected as the final reserve sites. Among them, eight coastal ports and seven cities belong to the flow nodes; inland ports and intersection make up only two. To some extent, these locations are consistent with China’s SPR reality by excluding Shanshan and Lanzhou. For scenario 2 (\(P \le 20,\;C \le 800\)), the results contain eight coastal ports and eight flow node cities; the number of both inland port and transit is two. However, compared with the current locations, the model does not select Shanshan, Lanzhou, Qingdao, Jinzhou, or Huizhou. For scenario 3 (\(P \le 25,\;C \le 1000\)), 21 cities are selected. Among them, five types of cities (coastal harbors, source, nodes, inland port, and transit) are nine, seven, seven, two, and three, respectively. The selection performs without Shanshan and Lanzhou and the scale. In scenario 4 (\(P \le 25,\;C \le 800\)), 23 cities are selected, including eight coastal harbors, nine source regions, nine source nodes, two inland ports, and four transits. However, Shanshan, Lanzhou, Qingdao, Jinzhou, and Huizhou do not feature in these results.
The results of scenarios 1 and 3 are superior to those of scenarios 2 and 4. Regarding the objective value, the maximum 10 million tonnes of storage capacity performs better than the maximum 8 million tonnes capacity. Therefore, for layout scheme matching, the scenario of the number of reserve sites being less than or equal to 25 and, meanwhile, the maximum capacity of reserve sites being 10 million tonnes provides the best the result.
Furthermore, 13 cities are selected in four scenarios, including Fuzhou, Dalian, Shanghai, Tianjin, Ningbo, Daqing, Xining, Xi’an, Dushanzi, Yueyang, Nanjing, Cangzhou, and Chengdu. Accordingly, the future construction work should mainly focus on these cities. In addition, concerning storage scales, Shanghai, Dalian, Tianjin, Ningbo, Xi’an, and Nanjing play an important role in reserving responsibility.
4.2.2 Model reliability and limitations
Model reliability can be verified because six cities in scenario 3 are in accordance with reality. Therefore, the results of scenario 3 should be adopted, taking full consideration of Tianjin, Qingdao, Dalian, Daqing, and Zhanjiang as future reserve sites. Meanwhile, Chengdu, Xi’an and Yueyang are expected to take reserve responsibility in China’s SPR Phase III.
In fact, the SPR storage site location model is limited by several assumptions and constraints. For example, each parameter may change (storage capacity, distance) due to economic, policy, and environmental influences, which can affect the accuracy location selection. However, considering the accessibility of domestic data, the model is only suited to a certainty situation. Generally, the emergency facility location model with uncertainty may estimate the locations more accurately. Although the SPR storage site location model takes into account the main influencing factors, the details of each reserve site, such as reserve type, reserve capacity, and future planning, are not discussed. Thus, policy makers should apply this model when more data are available to guide the SPR construction.
5 Conclusions
Determining optimal SPR storage site locations is an important and complicated problem. This study focuses on the problem of locating SPR storage sites in China. The location goals and influencing factors must be considered for resource accessibility, emergency efficiency, and a plausible spatial pattern. We have made assumptions which might overlook many features of China’s SPR construction. However, our application of the reserve site selection model generates many reasonable sites for future consideration. We believe that it can provide decision makers with a useful reference for Chinese SPR Phase III construction. The main results obtained for the SPR storage site location model are as follows:
-
(1)
The choice of SPR location is based on the minimum response time and minimum construction cost. The selection is based on the principles of justice, transparency, and efficiency. To achieve the goals, information about demand points and reserve sites and other related resources is explored for reserve site location determination.
-
(2)
Combined with the basic facilities location models, a multiple-objective programming model for SPR storage site location that satisfies the limitations of a set of constraints, such as reserve scale, storage capacity, and emergency periods, is introduced. Then, we design an algorithm based on Lingo to further improve the model.
-
(3)
According to the principle of proximity and the distribution of petroleum resources, the information on 52 demand points and 55 candidate reserve sites was collected. The optimal reserve site locations for China’s SPR vary under different scenarios. Specifically, the performance of 10 million tonnes of storage capacity is better than that of 8 million tonnes with the same reserve numbers; 13 cities are selected including Fuzhou, Dalian, Shanghai, Tianjin, Ningbo, Daqing, Xining, Xi’an, Dushanzi, Yueyang, Nanjing, Cangzhou, and Chengdu under each scenario. The scenario of number of reserve sites being less than or equal to 25 and, meanwhile, the maximum capacity of the reserve site being 10 million tonnes offers the best solution. Thus, policy makers should consider adopting the results of scenario 3 selecting Tianjin, Qingdao, Dalian, Daqing and Zhanjiang, Chengdu, Xi’an, and Yueyang to take the reserve responsibility in China’s SPR Phase III.
-
(4)
Although the built model has demonstrated its effectiveness for China’s SPR storage site determination, the influence from other uncertain factors should not be ignored. Researchers can further demonstrate the model’s feasibility by testing various uncertain factors.
References
Aouni B, Kettani O. Goal programming model: a glorious history and a promising future. Eur J Oper Res. 2001;133(2):225–31. doi:10.1016/S0377-2217(00)00294-0.
Bai Y, Zhou P, Tian L, et al. Desirable strategic petroleum reserves policies in response to supply uncertainty: a stochastic analysis. Appl Energy. 2016a;162:1523–9. doi:10.1016/j.apenergy.2015.04.025.
Bai Y, Zhou P, Zhou DQ, et al. Desirable policies of a strategic petroleum reserve in coping with disruption risk: A Markov decision process approach. Comput Oper Res. 2016b;66:58–66. doi:10.1016/j.cor.2015.07.017.
Chen HT. Location for strategic petroleum reserve base in China based on fuzzy comprehensive evaluation and maximal covering models. Sci Technol Manag Res. 2010;30(20):222–7 (in Chinese).
Chen H, Liao H, Tang BJ, et al. Impacts of OPEC’s political risk on the international crude oil prices: an empirical analysis based on the SVAR models. Energy Econ. 2016;57:42–9. doi:10.1016/j.ebeco.2016.04.018.
CNPC. Chinese refining industry under crude oil import right liberalization in 2015. https://eneken.ieej.or.jp/data/6405.pdf. Accessed 3 Nov 2015.
Current J, Min H, Schilling D. Multiobjective analysis of facility location decisions. Eur J Oper Res. 1990;49(3):295–307. doi:10.1016/0377-2217(90)90401-V.
Davis RM. National strategic petroleum reserve. Science. 1981;213(4508):618–22. doi:10.1126/science.213.4508.618.
De Vos NJ, Rientjes THM. Multi-objective training of artificial neural networks for rainfall-runoff modeling. Water Resour Res. 2008. doi:10.1029/2007WR006734.
Dong X, Zhou Z, Li H. Improve the government strategic petroleum reserves. Adv Chem Eng Sci. 2013. doi:10.4236/aces.2013.34A1001.
Dong KY, Sun RJ, Li H, et al. A review of China’s energy consumption structure and outlook based on a long-range energy alternatives modeling tool. Pet Sci. 2017;14:214–27. doi:10.1007/s12182-016-0136-z.
Fan Y, Zhang XB. Modelling the strategic petroleum reserves of China and India by a stochastic dynamic game. J Policy Model. 2010;32(4):505–19. doi:10.1016/j.jpolmod.2010.05.008.
Farahani RZ, Asgari N, Heidari N, et al. Covering problems in facility location: a review. Comput Ind Eng. 2012;62(1):368–407. doi:10.1016/j.cie.2011.08.020.
Helm D. Energy policy: security of supply, sustainability and competition. Energy Policy. 2002;30(3):173–84. doi:10.1016/S0301-4215(01)00141-0.
Hubbard RG, Weiner RJ. Managing the strategic petroleum reserve: energy policy in a market setting. Annu Rev Energy. 1985;10(1):515–56. doi:10.1146/annurev.eg.10.110185.002503.
Jiao JL, Han KY, Wu G, et al. The effect of an SPR on the oil price in China: a system dynamics approach. Appl Energy. 2014;133:363–73. doi:10.1016/j.apenergy.2014.07.103.
Karl TL. Oil-led development: social, political, and economic consequences. Encycl Energy. 2007;4:661–72. doi:10.1016/B0-12-176480-X/00550-7.
Li WL, Tan JH. Study of the storage of strategic petroleum reserves in China. China Offshore Oil Gas Eng. 2002;14(3):9–14 (in Chinese).
Li X, Zhao Z, Zhu X, et al. Covering models and optimization techniques for emergency response facility location and planning: a review. Math Methods Oper Res. 2011;74(3):281–310. doi:10.1007/s00186-011-0363-4.
Liu MZ, Qu CZ, Feng YF. Development and application of an optimal layout model for national coal emergency reserve. China Saf Sci J. 2014;8:024 (in Chinese).
Majid NDA, Shariff AM, Loqman SM. Ensuring emergency planning & response meet the minimum process safety management (PSM) standards requirements. J Loss Prev Process Ind. 2016;40:248–58. doi:10.1016/j.jlp.2015.12.018.
Martínez-Palou R, de Lourdes Mosqueira M, Zapata-Rendón B, et al. Transportation of heavy and extra-heavy crude oil by pipeline: a review. J Pet Sci Eng. 2011;75(3):274–82. doi:10.1016/j.petrol.2010.11.020.
Mladenović N, Brimberg J, Hansen P, et al. The p-median problem: a survey of metaheuristic approaches. Eur J Oper Res. 2007;179(3):927–39. doi:10.1016/j.ejor.2005.05.034.
Niu YJ, Guo JK, Shao HY. Study of the location evaluation of national strategic depots for refined oil products. Logist Technol. 2010;29(16):149–50 (in Chinese).
Niu YJ, Mu X, Zhao CJ, et al. Evaluation of site selection for national strategic reserve depots of refined petroleum products. Adv Mater Res. 2013;779:1607–12. doi:10.4028/www.scientific.net/AMR.779-780.1607.
Oregon Department of Transportation Research Unit. Selection criterial for using nighttime construction and maintenance operations. http://www.oregon.gov/ODOT/TD/TP_RES/ResearchReports/SelCritNighttimeCon.pdf. Accessed May 2003.
Park YS. China’s energy security strategy: implications for the future Sino-US relations. Int J Soc Sci Stud. 2015;3(2):30–40. doi:10.11114/ijsss.v3i2.670.
Roth R. Computer solutions to minimum-cover problems. Oper Res. 1969;17(3):455–65. doi:10.1287/opre.17.3.455.
The Oxford Institute for Energy Studies. The structure of China’s oil industry: past trends and future prospects. 2016. https://www.oxfordenergy.org/wpcms/wp-content/uploads/2016/05/The-structure-of-Chinas-oil-industry-past-trends-and-future-prospects-WPM-66.pdf. Accessed May 2016.
Tillerson JR. Geomechanics investigations of SPR crude oil storage caverns. Albuquerque: Sandia Labs; 1979.
Toregas C, Swain R, ReVelle C, et al. The location of emergency service facilities. Oper Res. 1971;19(6):1363–73. doi:10.1287/opre.19.6.1363.
Williams JC. Optimal reserve site selection with distance requirements. Comput Oper Res. 2008;35(2):488–98. doi:10.1016/j.cor.2006.03.012.
Wilson DT, Hawe GI, Coates G, et al. A multi-objective combinatorial model of casualty processing in major incident response. Eur J Oper Res. 2013;230(3):643–55. doi:10.1016/j.ejor.2013.04.040.
Wu K. China’s energy security: oil and gas. Energy Policy. 2014;73:4–11. doi:10.1016/j.enpol.2014.05.040.
Wu P, Wang XZ. Description of China’s strategic oil reserves. Translated from China Oil Weekly. http://hexun.com/2011-01=17/126848881.html. Accessed on 2 May 2012.
Zhang T, Lv XD, Zhang YF. Location-allocation problems of strategic petroleum storage based on a hybrid intelligent algorithm. Oil Gas Storage Transp. 2008;27(10):12 (in Chinese).
Zhu YC. Research on the position for strategic petroleum reserve. Future Dev. 2007;10:8–11 (in Chinese).
Acknowledgments
We gratefully acknowledge that this work was supported by the National Natural Science Foundation of China (Nos. 71273277/71373285/71303258) and the Philosophy and Social Sciences Major Research Project of the Ministry of Education (No. 11JZD048). Helpful comments by anonymous reviewers are most appreciated.
Author information
Authors and Affiliations
Corresponding author
Additional information
Edited by Xiu-Qin Zhu
Appendices
Appendix 1
See Table 8.
Appendix 2: Lingo programming
sets: |
a/1..52/:w;!i; |
b/1..55/:y;!j; |
link1(a,b):d,x;!; |
endsets |
data: |
d=@ole(‘d.xls’);! Shortest path matrix; |
w=@ole(‘w.xls’);! Amount of each demand point; |
enddata |
min=L |
@for(a(i):@sum(link1(i,j):d(i,j)*x(i,j))<=L); |
@for(a(i):@for(b(j):x(i,j)<=y(j))); |
@for(a(i):@sum(b(j):x(i,j))=1); |
@sum(b(j):y(j))<=25; |
@for(b(j):@sum(a(i):w(i)*x(i,j))<=y(j)*1000); |
@for(b(j):@sum(a(i):w(i)*x(i,j))>=y(j)*100); |
@for(a(i):@sum(link1(i,j):d(i,j)*x(i,j))<=900); |
@for(link1(i,j):@bin(x(i,j))); |
@for(b(j):@bin(y(j))); |
End |
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Li, H., Sun, RJ., Dong, KY. et al. Selecting China’s strategic petroleum reserve sites by multi-objective programming model. Pet. Sci. 14, 622–635 (2017). https://doi.org/10.1007/s12182-017-0175-0
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12182-017-0175-0