In the context of a rapidly evolving global economic landscape and an increasingly complex market environment,the demand for mineral resources has become highly unpredictable.This unpredictability poses significant challenges to the stable operation and optimization of ore supply chain networks.Consequently,the rational design of ore transportation network structures and the optimization of ore flow have emerged as critical issues necessitating urgent attention within the domain of ore supply chain network optimization.To address the complexities associated with optimizing ore supply chains under uncertain demand conditions,this study undertook an in-depth exploration and introduced innovative solutions.A novel robust possibility fuzzy programming model was developed,incorporating considerations of carbon taxes and carbon emission limits.During the model construction process,the uncertainties in ore demand and transportation costs were comprehensively represented using trapezoidal fuzzy numbers.By employing the robust possibility fuzzy programming method,the complex uncertainty problem was transformed into a solvable deterministic form,facilitating the development of a model aimed at minimizing total operating costs.This study establishes a scientifically robust and effective quantitative framework for cost management within the ore supply chain.To efficiently address the proposed model,a differential evolution algorithm incorporating utility degree ranking was introduced.This algorithm integrates the COPRAS model,thereby overcoming the limitations of traditional algorithms that rely exclusively on fitness functions.It assesses the benefits of the ore supply chain across multiple dimensions,including economic costs,environmental impacts,and resource utilization efficiency.By employing utility degree-based population ranking,the algorithm directs iterative processes towards optimal solutions,significantly enhancing its search capabilities and convergence speed.To validate the effectiveness and superiority of the proposed model and algorithm,they were applied to real-world engineering scenarios.The results demonstrated that,in comparison to the traditional differential evolution algorithm based on fitness functions,the utility degree-based differential evolution algorithm achieved superior solution efficiency with substantially reduced computational time.Regarding performance,the proposed approach demonstrated accelerated convergence towards optimal solutions and exhibited more significant optimization effects on the ore supply chain network.Moreover,when compared to conventional meta-heuristic algorithms like genetic algorithms,the utility degree-based differential evolution algorithm outperformed in terms of utility degree metrics.It required fewer iterations to achieve convergence and exhibited enhanced global search capabilities and optimization potential.Consequently,this research offers a solid theoretical foundation and practical technical support for ensuring the efficient operation and sustainable development of ore supply chains within complex and dynamic market environments.
矿业作为经济社会发展的基石,肩负着高效绿色开发和清洁低碳利用矿产资源的重任(王家臣等,2024)。在全球经济一体化和市场环境动态变化的背景下,金属矿产资源需求的不确定性愈发显著,这使得矿石供应链的高效可持续运营成为矿业发展的核心诉求(Leal Gomes Leite et al.,2019;Zhang et al.,2024)。矿石供应链涵盖从矿山开采到终端产品制造的复杂流程,其中基础设施选址、作业协同和矿石流配置等环节均对矿石供应链整体运营效率和稳定性具有关键影响(牛占奎等,2016;Zeng et al.2021;Zhang et al.,2024)。因此,在需求不确定的商业环境下,优化矿石供应链网络结构,尤其是合理配置运输网络和优化矿石流运输,已成为亟待解决的关键问题。
不确定性广泛存在于矿石供应链运营中,需求等参数的不确定导致所构建的矿石供应链网络优化模型不可解。因此,采用不确定建模技术对上述优化模型进行确定性等价模型推导。鲁棒可能性模糊规划已被证实为处理混合不确定性下优化问题的有效方法,并被广泛应用于供应链网络优化领域(Zhang et al.,2009;Yousefi et al.,2017;Ghasemi et al.,2022)。基于问题描述,模型包含的不确定参数为矿石需求和运输成本。首先,将1.2小节中矿石供应链网络结构模型简化为
目标函数中,向量表示不随情景改变的固定成本参数,向量表示情景可变参数。约束条件中,向量、和为资源系数,和为情景可变限额系数,为不随情景改变的固定资源限额系数,为等式约束的资源系数。为随不确定情景变化的连续变量,为二元变量。可能性机会约束规划中,目标函数的期望值用于描述不确定性(Quddus et al.,2018;Borajee et al.,2023)。此外,不确定参数在不同需求情景下通常建模为模糊数,包括三角模糊数和梯形模糊数。相比三角模糊数,梯形模糊数刻画不确定参数,能够消除不确定性。因此,模型采用梯形模糊数用于刻画不确定参数,并得出如下可能性机会约束规划模型:
式中:不确定参数采用梯形模糊数,表示;和分别为不确定参数的满足度水平,通常根据决策者的决策偏好得到(Nayeri et al.,2020),取0和1之间。 基于风险决策理论,根据决策者的风险偏好,可将决策者划分为保守型、中间型和冒险型。满足度水平越大,意味着决策者的风险态度偏向冒险型,对于约束条件的限制更加严格,以约束条件式(1)为例,右端项越大,约束条件越严格;满足度水平越小,意味着决策者应对风险更加保守,对于约束条件的限制较放松。然而,在情景不确定条件下,目标函数对期望值的偏差敏感,意味着可能性机会约束规划模型不能得到可靠的鲁棒解。情景鲁棒优化通过在目标函数中引入惩罚函数和期望函数,从而消除不确定参数对结果的影响,保证情景不确定下解和模型的鲁棒性。
BelovG, BolandN L, SavelsberghM W P,et al,2020.Logistics optimization for a coal supply chain[J].Journal of Heuristics,26(2):269-300.
[2]
BorajeeM, Tavakkoli-MoghaddamR, Madani-SaatchiS H,2023.A chance-constraint optimization model for a multi-echelon multi-product closed-loop supply chain considering brand diversity:An accelerated Benders decomposition algorithm[J].Computers and Operations Research,152:106130.
[3]
ChaharmahaliG, GhandalipourD, JasemiM,et al,2022.Modified metaheuristic algorithms to design a closed-loop supply chain network considering quantity discount and fixed-charge transportation[J].Expert Systems with Applications,202:117364.
[4]
ChengQiang, GaoYuanjie, ChuHongyan,et al,2023.Flexible job shop scheduling of machining based on multi- objective differential evolution algorithm[J].Journal of Beijing University of Technology,49(3):335-345.
[5]
DongHai, GaoXiuxiu, WeiMingqi,2023.Modeling and optimization solution of two-stage emergency supply chain network under uncertain environment[J].Chinese Journal of Management Science,31(12):107-116.
[6]
GhasemiP, GoodarzianF, AbrahamA,et al,2022.A possibilistic-robust-fuzzy programming model for designing a game theory based blood supply chain network[J].Applied Mathematical Modelling,112:282-303.
[7]
GoodarzianF, Hosseini-NasabH, MuñuzuriJ,et al,2020.A multi-objective pharmaceutical supply chain network based on a robust fuzzy model:A comparison of meta-heuristics[J].Applied Soft Computing,92:106331.
[8]
Kusi-SarpongS, SarkisJ, WangX P,2016.Assessing green supply chain practices in the Ghanaian mining industry:A framework and evaluation[J].International Journal of Production Economics,181:325-341.
[9]
LaRoche-BoisvertM, DimitrakopoulosR, FerlandJ A,2021.Simultaneous production scheduling and transportation optimization from mines to port under uncertain material supply[J].Resources Policy,73:102150.
[10]
Leal Gomes LeiteJ M, ArrudaE F, BahienseL,et al,2020.Modeling the integrated mine-to-client supply chain:A survey[J].International Journal of Mining,Reclamation and Environment,34(4):247-293.
[11]
LinJian, YeJingxuan, LiuWenwen,et al,2023.Multimodal differential evolution algorithm for solving capacitated vehicle routing problem[J].Journal of Computer Applications,43(7):2248-2254.
[12]
LiuW B,2021.A robust optimization modeling for mine supply chain planning under the big data[J].Wireless Communications and Mobile Computing,(1):1709363.
[13]
MaYinyu,2023.Technological and economic on optimization of underground mine transportation route based on ant colony optimization[J].Modern Mining,39(5):250-254.
[14]
NayeriS, PaydarM M, Asadi-GangrajE,et al,2020.Multi-objective fuzzy robust optimization approach to sustainable closed-loop supply chain network design[J].Computers and Industrial Engineering,148:106716.
[15]
NiuZhankui, LiWenxin,2016.Main risks and control strategies of supply chain logistics enterprises:Taking Huai mine logistics as an example[J].Marketing Management Review,(10):44-45.
[16]
QuddusM A, ChowdhuryS, MarufuzzamanM,et al,2018.A two-stage chance-constrained stochastic programming model for a bio-fuel supply chain network[J].International Journal of Production Economics,195:27-44.
[17]
ShiShuangyuan, XiongHegen,2024.Multi-objective differential evolution algorithm for no-tardiness job shop scheduling problem with outsourcing option[J].Computer Integ-rated Manufacturing Systems,30(12):4352-4368.
[18]
SinghG, ErnstA T, BaxterM,et al,2015.Rail schedule optimisation in the hunter valley coal chain[J].RAIRO-Operations Research,49(2):413-434.
[19]
ValderramaC V, Santibanez-GonzálezE, PimentelB,et al,2020.Designing an environmental supply chain network in the mining industry to reduce carbon emissions[J].Journal of Cleaner Production,254:119688.
[20]
WangJiachen, LiuYunxi, LiYang,et al,2024.60 years development and prospect of mining systems engineering[J].Journal of China Coal Society,49(1):261-279.
[21]
WangYadong, ShiQuan, ZhangFang,et al,2019.Multi-stage spare parts supply optimization based on dynamic evolutionary algorithm[J].Systems Engineering and Electronics,41(11):2514-2523.
[22]
YousefiS, SoltaniR, Farzipoor SaenR,et al,2017.A robust fuzzy possibilistic programming for a new network GP-DEA model to evaluate sustainable supply chains[J].Journal of Cleaner Production,166:537-549.
ZengL Y, LiuS Q, KozanE,et al,2023.Designing a resilient and green coal supply chain network under facility disruption and demand volatility[J].Computers and Industrial Engineering,183:109476.
[25]
ZhangQ, LiuS Q, D’ArianoA,et al,2024.A bi-level programming methodology for decentralized mining supply chain network design[J].Expert Systems with Applications,250:123904.
[26]
ZhangX D, HuangG H, NieX H,2009.Robust stochastic fuzzy possibilistic programming for environmental decision making under uncertainty[J].Science of The Total Environment,408(2):192-201.
[27]
ZhouH N, SamavatiM, HillA J,2021.Heuristics for integrated blending optimisation in a mining supply chain[J].Omega,102:102373.