Optimal Distribution Model Research Articles

Estimating Ruin Probability in an Insurance Risk Model Using the Wang-PH Transform Through Claim Simulation

The accurate estimation of ruin probability is a fundamental challenge in non-life insurance, impacting financial stability, risk management strategies, and operational decisions. This study aims to propose an approach for estimating ruin probability using claim simulation enhanced by the Wang-PH transform to fit various loss distributions, including Gamma, Weibull, Lognormal, Log-logistic, Inverse Weibull, and Inverse Gaussian, to actual claim data. Methods involve the transformation of loss distributions via the Wang-PH transform and rigorous evaluation to select the optimal distribution model that best reflects actual claim characteristics. This model serves as the foundation for estimating finite-time ruin probability through claim simulation, employing the acceptance-rejection technique to generate random samples. Additionally, a regression-based methodology estimates the minimum capital reserve required to safeguard against financial risk. Findings indicate the proposed method's computational efficiency, making it a valuable tool for insurers and risk analysts in assessing and mitigating financial risks in the non-life insurance sector. The novelty of this study lies in the integration of the Wang-PH transform with empirical data fitting and simulation techniques, applied to estimating ruin probability and determining capital reserves. Doi: 10.28991/ESJ-2025-09-01-011 Full Text: PDF

AI-Driven Transport and Distribution Optimization Model (TDOM) for the downstream petroleum sector: enhancing sme supply chains and sustainability

The downstream petroleum sector faces significant challenges in transport and distribution logistics, particularly for small and medium enterprises (SMEs) that rely on efficient supply chains to remain competitive. Inefficiencies such as fluctuating demand, fuel shortages, and route optimization hurdles contribute to increased operational costs and environmental impact. Addressing these persistent issues requires innovative solutions that balance efficiency, cost reduction, and sustainability. This study introduces the AI-Driven Transport and Distribution Optimization Model (TDOM), a transformative approach designed to revolutionize logistics in the downstream petroleum sector. TDOM leverages advanced machine learning algorithms and predictive analytics to optimize delivery routes, reduce fuel consumption, and ensure real-time adaptability to market demands. By integrating data from diverse sources, including traffic patterns, fuel station inventories, and weather conditions, TDOM provides SMEs with actionable insights for enhanced decision-making. The primary objectives of this research are threefold: to improve supply chain efficiency for SMEs in the downstream petroleum sector, to minimize environmental impacts through reduced carbon emissions, and to ensure cost-effective transport and distribution processes. The methodology includes developing a robust AI-driven model utilizing supervised and unsupervised learning techniques, followed by simulation testing on real-world logistics data. Key performance metrics such as delivery time, cost savings, and carbon footprint reduction will be analyzed to validate the model's efficacy. Anticipated outcomes of the TDOM implementation include a 20-30% reduction in transport costs, a significant decrease in fuel consumption, and improved supply chain resilience for SMEs. Furthermore, the model is expected to enhance sustainability by promoting eco-friendly logistics practices, thereby aligning with global environmental standards. By addressing inefficiencies in downstream petroleum logistics and supporting SMEs, the AI-driven TDOM represents a paradigm shift in how the industry approaches transport and distribution challenges. This innovative solution underscores the potential of artificial intelligence to transform traditional industries and foster a more sustainable future.

Research on Cross-Border e-Commerce Supply Chain Prediction and Optimization Model Based on Convolutional Neural Network Algorithm

Enhancing the precision of supply chain management and reducing operational costs are crucial for the development of the cross-border e-commerce market. However, existing research often overlooks the demand uncertainty caused by seasonal variations and the challenges of handling returns in logistics. Therefore, this paper proposes a SARIMA-CNN-BiLSTM prediction model that effectively captures both the seasonal and nonlinear characteristics of cross-border e-commerce supply chains. Additionally, by incorporating the returns process, a supply chain distribution optimization model is developed with the objective of minimizing total operational costs. The model is solved using an improved whale optimization algorithm. In validation with real-world data, the SARIMA-CNN-BiLSTM model achieved a mean absolute percentage error reduction of 6.479 and 7.703 compared to convolutional neural network (CNN) and BiLSTM models, respectively. Moreover, the chosen optimization algorithm reduced the cost by 231,310 CNY, 62,564 CNY, and 131,632 CNY compared to the whale optimization algorithm, genetic algorithm, and particle swarm optimization, respectively. The proposed approach provides robust support for cross-border e-commerce enterprises in reducing costs and enhancing efficiency in their operations.

Optimization of airflow distribution in mine ventilation networks considering ventilation energy consumption and number of regulators

In view of the problem of high energy consumption and high control costs caused by uneven airflow distribution and unreasonable control in complex mine ventilation networks, this study takes the minimum ventilation energy consumption and number of regulators as optimization objectives to establish a multi-objective optimization model for airflow distribution in mine ventilation networks. Based on the roadway adjustable attributes and the minimum spanning tree principle, the location of regulators was reasonably determined. Moreover, this article proposes an improved invasive weed optimization (IIWO) algorithm to solve the optimization model with high coupling and nonlinearity. Compared with other algorithms, IIWO showed excellent optimization performance. IIWO was applied to optimize the airflow distribution in the mine ventilation network. The results show that the algorithm can effectively reduce the energy consumption and number of regulators of the ventilation network, and the energy saving rate of the fan is 31.78%.

Методи та засоби оптимізації використання обчислювальних ресурсів в корпоративній мережі закладу вищої освіти

The paper considers the problems of using computing resources in the corporate network of a higher education institution. An approach to assessing the efficiency of the load of computer workstations is proposed, which is based on the indicators of the load of individual computing resources and their implementation in the general system of computing resources for individual applied resource-intensive tasks. It is established that to minimize costs when building a university CM, it is necessary to use non-standard approaches that could dynamically distribute computing resources, in particular terminal and cluster technologies for building corporate networks. A number of experimental studies have been conducted based on data on the functioning of the computer network of a higher education institution. A mathematical model of optimal load distribution on terminal servers has been constructed, which describes the dependence between individual network nodes and the file access system.

Multi-charging Station access distribution network grid planning and optimization method based on MRFO algorithm

Abstract The traditional multi-charging station integration into the distribution network planning and optimization model is a regional optimization process with low overall efficiency, increasing the total optimization cost. Therefore, a multi-charging station access distribution network based on the MRFO (Manta Ray Foraging Optimization) algorithm is proposed. The MRFO calculation method was adopted to improve overall efficiency, and the design of the MRFO calculation multi-charging station access distribution network planning optimization model was completed. Optimization processing and multi-objective validation are used to achieve grid optimization. The test results show that the MRFO calculation and distribution network planning optimization method designed in this study ultimately achieve a relatively low total optimization cost, better optimization effect, and practical application value.

Integrating heterogeneous information for modeling non-stationarity of extreme precipitation in the Yangtze River Basin

Extreme precipitation exhibits non-stationary characteristics due to climate change and land use/cover change in recent years. In the literature, probability distributions with varying parameters associated with covariates have been applied to describe the non-stationarity of extreme precipitation. However, there is no consensus on the optimal distributions and the number and type of covariates for constructing non-stationary models. This study comprehensively investigates and determines the optimal distribution models, the type of covariates, and the number of covariates to best characterize the non-stationarity of extreme precipitation, using the Yangtze River Basin as the study area. The time series of four indices representing extreme precipitation of different durations and intensities are fitted by stationary and non-stationary Generalized Additive Models for Location, Scale, and Shape. The models are constructed using seven distributions and four types (time, global climate, local climate, and land cover) of covariates. Each model incorporates one to three covariates simultaneously. The results show that the most preferred distribution is the Generalized Extreme Value distribution for very wet-day precipitation, while the Log-Normal distribution is more suitable for other extreme precipitation indices. For the type of covariates, when using a single covariate, the proportion of stations using a global climate covariate as the best covariate (22.3 %–33.2 % of 193 stations) is close to that using a local climate covariate as the best covariate (22.3 %–27.5 %). The proportion of stations using land cover as the best covariate is much lower (7.3 %–10.9 %), while that using time as the best covariate is the lowest (1.6 %–4.1 %). When using multiple covariates, the models with both global and local climate covariates are mostly selected (8.3 % to 17.6 %) for all extreme precipitation indices. For the number of covariates, one or two covariates are enough to describe the non-stationary behaviors of all extreme precipitation indices. This study emphasizes the necessity to use non-stationary models for extreme precipitation frequency analysis with careful consideration and selection of distributions and covariates.

Deriving rainfall IDF curves using modified Bartlett-Lewis rectangular pulses (BLRP) model for Babylon City, Iraq

In this study, daily rainfall data spanning 32 years, from 1991 to 2022, were collected from the Babylon City Meteorological Station (BCMS). The modified Bartlett-Lewis Rectangular Pulses (BLRP) model with seven parameters was used to generate the intensity-duration-frequency (IDF) curve for durations of 2, 5, 10, 25, 50, and 100 years. The estimated extreme values obtained from BCMS were fitted with EasyFit software using three statistical continuous distributions: the Gumbel (EV-1), Generalized Extreme Value (GEV), and Log-Pearson type III (LP-3) to determine the optimal distribution model for the IDF curve. The AIC and BIC criteria were employed to rank the generated IDF curves for each duration to select the optimal distribution model. The results of the AIC criteria ranked the PDFs as LP-3 first model, EV-1 second model, and GEV third model with sums of weighted rank values as 16, 17, and 27, respectively. In contrast, the BIC criteria ranked the PDFs as EV-1 first model, LP-3 second model, and GEV third model with sums of weighted rank values as 16, 17, and 27, respectively. The EV-1 and LP-3 models were selected as the most suitable results for simulating the maximum rainfall in Babylon City for various return periods. This decision was based on the total score of the special rankings of the two tests, with EV-1 and LP-3 having the lowest values of AIC and BIC.

A cost-efficient content distribution optimization model for fog-based content delivery networks

The massive data demand requires content distribution networks (CDNs) to use evolving techniques for efficient content distribution with guaranteed quality of service (QoS). The distributed fog-based CDN model, with optimal fog node placements, is a suggested aproach by researchers to meet this demand. While many studies have focused on improving QoS by optimizing fog node placement, they have rarely considered the impact on content distribution, affected by placement, usage changes, and delivery rates. Therefore, the practical approach to fog node placement for CDN services must examine its impact on content distribution. Further, current research on fog-based CDN lacks formal methods to address key challenges: R1) strategic placement of fog nodes to process end-user requests; R2) construction of a content distribution path with guaranteed QoS; R3) cost minimization of building a fog-based CDN model. We construct this as a joint optimization problem by considering four parameters: geographical regions, open public Wi-Fi access points (OPWAPs) locations, QoS, and cost to achieve research objectives R1–R3. As a solution, we propose a dual-step framework. First, a heuristic for optimal fog node placement based on geographic regions and OPWAP locations is proposed. Second, we propose two algorithms, Greedy Performance-based Node Selection (GPDS) and Greedy Fog Node Selection algorithm (GFNSA), for selecting fog nodes, minimizing the cost of building a fog-based CDN while achieving optimal content distribution paths. The results demonstrate that the proposed methods outperform the baseline techniques and provide near-optimal solutions to the problem.

Probability Distribution Analysis of Annual Maximum Precipitation in the Niger Delta: A Critical Step towards Effective Flood Risk Management

Aims: The study aimed to analyze historical maximum precipitation data from seven cities in the Niger Delta to assess vulnerability to extreme rainfall events, identify optimal probability distribution models for future predictions, and provide insights for flood risk management strategies. Study Design: This is a retrospective analytical study based on historical precipitation data. Place and Duration of Study: The study was conducted across seven major cities in the Niger Delta region of Nigeria, utilizing data collected over several decades. Methodology: Historical maximum precipitation data were statistically analyzed to determine the best-fit probability distribution models. The Kolmogorov-Smirnov (KS) test and Mean Squared Prediction Error (MSPE) were used to validate the suitability of distributions, including Log-Normal, Normal, Log-Pearson III, and Generalized Extreme Value (GEV) for each city. Return periods of 10, 25, 50, and 100 years were calculated to predict future precipitation trends and assess flood risks. Results: Significant variations in annual maximum precipitation were observed across the cities, with Calabar recording the highest peak at 4062.7mm. The Log-Normal distribution was the best fit for Akure and Calabar, while the Normal distribution best described rainfall in Benin City. Log-Pearson III was optimal for Owerri, Umuahia, and Uyo, and GEV best fitted Port Harcourt’s data. High P-values (>0.87) indicated good model fits across the cities. Return period analysis suggested greater risks of extreme precipitation events in coastal cities like Calabar and Port Harcourt. Conclusion: The study underscores the importance of tailored, city-specific flood management strategies in the Niger Delta to mitigate the impact of extreme rainfall events, particularly in the face of climate change.

Optimization on distribution of high thermal conductivity materials in metal hydride reactor for improving heat transfer performance

The poor heat transfer performance significantly limits the reaction rate of hydrogen storage and release in metal hydrides reactors. In this work, the expanded graphite was used as high thermal conductivity materials (HTCMs), its optimal distribution in metal hydrides reactor was obtained by topology optimization method to enhance the heat transfer performance. Comparative studies on the hydrogen storage and release performance were conducted for the optimal HTCMs distribution model, uniform HTCMs distribution model, reconstructed model, and unimproved model to validate the optimization results. Finally, the effect of HTCMs contents was also studied. The results show that the optimization results shorten the hydrogen storage and release time by 32.3 % and 15.9 %, respectively, compared to the uniform distribution model. Besides, the proposed optimization scheme exhibits good performance under different HTCMs contents. When the content was set to 12 %, the optimized model reduces the hydrogen storage and release time by 28.5 % and 10.8 %, compared to the results of uniform distribution. The method and results of topology optimization have high reliability, and the HTCMs distribution designed by topology optimization has better hydrogen storage and release performance compared to original designs, it is widely applicable.

A two-stage optimization model for relief logistics and distribution to disaster survivors under two-fold uncertainty

Disasters are unforeseen occurrences requiring extensive transport deployment to support and relieve victims. Sometimes, this transportation is not feasible directly from some supply points to some destination points. Due to this tragedy, it is unclear precisely what is available at supply points, what is needed at destinations, how much transportation capacity there is, and what the routes are like. In this study, we investigate a two-stage multi-item fixed charge four-dimensional transportation problem using the concept of big data theory under the two-fold uncertainties. Here, the model’s parameters such as unit transportation costs, availabilities of items at the suppliers, fixed charges, capacities of conveyances, and demands of the items at the retailers are considered type-2 zigzag uncertain variables. Using big data theory and based on uncertain programming theory, two novel uncertain models are developed such as chance-constrained programming and expected value programming model. These two uncertain models transformed into the deterministic form via uncertainty inverse distribution theory. A critical value based reduction method with three categories (i.e., expected value, pessimistic value, and optimistic value) is applied to reduce the type-2 zigzag uncertain variable to the type-1 zigzag uncertain variable. The genetic algorithm and particle swarm optimization techniques have been proposed to find the optimal solution for the two deterministic models. The efficiency of our proposed approach is demonstrated with a real-life numerical example.

Optimum management of microgrid generation containing distributed generation sources and energy storage devices by considering uncertainties

In recent years, the financial and energy crises have made the economical and safe operation of distribution microgrids one of the main challenges for operators. In this regard, the optimal distribution of the microgrid, considering the favorable indicators of Generation and consumption, significantly improves the quality of the development and economic operation of electricity grids. On the other hand, distribution microgrids have many problems and complications due to the presence of DG sources and the uncertainties caused by them in Generation and consumption, such as the inability to solve optimal load distribution using common and common calculation methods. Therefore, grid technical indicators such as voltage deviation and losses in solving load distribution add to the complexity of related issues. To solve the challenges raised in the distribution microgrid, this paper proposes an optimal load distribution model that models the uncertainties in the distribution microgrids using the Monte Carlo method. Using meta-heuristic algorithms, the optimal Generation function by modeling grid security constraints such as voltage deviation, losses, and a load limit of the system should be in economic operation mode, and then the power distribution equations should be solved. The battery can supply 1 kW of active power to the system at a cost of 1 unit/kW during the high cost period if it absorbs 1 kW of active power at a cost of 0.92 units/kW during the low cost period. Additionally, the active and reactive power losses have been decreased to 0.3kW and 0.2kW, respectively, by utilizing the suggested methodology. The simulation results show that the proposed method performs satisfactorily and that the modeling is simulated in MATLAB software.

Reinforcement learning control of multi-spectral LED systems for rendering optimal lighting performance in indoor environment

Light-emitting diodes (LEDs) have superseded traditional light sources due to their superior qualities. However, achieving desired spectral distribution in indoor environment still remains a challenge, particularly for optimizing multiple lighting parameters, which are nonlinear functions of the light spectra, including correlated color temperature (CCT), color rendering index (CRI) and circadian action factor (CAF). Existing studies have been mainly devoted to seeking optimal solutions, by means of various optimization methods, e.g. genetic algorithms. However, these methods can only result in fixed recipes to achieve pre-assigned reference parameters, and cannot meet the need of dynamic varying lighting environment. Therefore, this paper proposes a novel framework for optimal spectrum design and control in multichannel LED lighting systems. It surpasses traditional methods by employing reinforcement learning (RL) to generate an entire optimal spectral power distribution (SPD) model, comprehensively considering CCT, CRI, CAF, and other relevant parameters. This framework integrates the RL-based optimal SPD generator with a corresponding real-time controller, bridging the gap between simulations and real-world applications. The effectiveness of this new framework has been verified by an experimental prototype, showcasing both the optimal SPD and the achieved control performance. This work paves the way for high-quality, user-centric lighting with dynamic CCT and spectral control, and enables the creation of lighting environments that optimize human well-being, productivity, and overall experience.

Performance optimization and energy minimization of cloud data center using optimal switching and load distribution model

Cloud computing is an effective computing methodology used in all stages of business. Most of the Cloud Data Centers (CDC) operates on the basis of peak load and huge scales. Hence, it necessitates saving the energy in CDC. This study introduces an energy-efficient strategy based on the fat tree. Here, Taylor-based Manta-Ray Foraging Optimization (Taylor-MRFO) is developed by combining the Taylor series with Manta Ray Foraging Optimization (MRFO) to distribute the load in a CDC. In load distribution, the cloud data switching to the preferred mode is done by the Actor critic neural network (ACNN). Furthermore, the developed Taylor-MRFO+ACNN provided a better outcome than the conventional approaches with the least energy consumption of 0.4930, least load of 0.3631, and least fitness of 0.4343. For setup-1, when the population size is 15, the load value obtained by the proposed method is 23.43 %, 10.19 %, 7.18 %, 5.31 %, 4.43 %, and 2.58 % higher when compared to the existing approaches namely, Artificial Bee colony(ABC), Efficient Load Optimization and Resource Minimization (ELORM), Adaptive Parameter- Ant Colony Optimization (AP-ACO), Multi-Objective Memetic Algorithm-Adaptive Plant Intelligent Behavior Optimization (MOMA-APIBO), Cooling Control Algorithm (CCA), and Minimum Total Power (MinPR).

Integrated production simulation of renewable energy and transportation consumption

Abstract In order to improve the reserve adequacy of high proportion new energy power systems and realize the efficient consumption of large-scale new energy, the source of reserve resources is expanded to the source network and load storage, and the operating reserve level that can be provided by the source network and load storage flexible regulating resources is quantified in terms of the direction of regulation and the mode of participation. Then, according to the regulation idea of “synchronous units bear the responsibility during regular hours and multiple types of resources share the responsibility during special hours”, a coordinated and participatory operating reserve capacity allocation strategy of source, network, load, and storage is proposed, and a reserve distribution optimization model incorporating multiple types of heterogeneous resources is constructed to achieve the optimization of reserve distribution of each type of reserve resource in the whole network. The optimization model of standby distribution is constructed to incorporate multiple types of heterogeneous resources and realize the step-by-step refinement of standby resources distribution among the whole grid - resources of the same type - units/stations. A case study is carried out based on a provincial power grid, and the effectiveness of the proposed strategy is verified by comparing it with the existing operational reserve allocation guidelines.

Robust optimization model for relief supplies distribution considering fairness

AbstractThe emergency management agency (EMA) needs to distribute limited relief supplies efficiently. However, it is difficult to develop a reliable system for relief supplies distribution owing to the uncertainties in emergencies. This paper investigates a robust multiperiod relief supplies distribution problem that considers the uncertainties of transportation time, the amount of donation amount, and the secondary disasters. First, a satisfaction model is constructed by considering the relief supplies and the transportation time. The negative effect of the delay in transportation time is considered in the satisfaction model. Second, based on the satisfaction model, a comprehensive fairness strategy is constructed that considers both vertical fairness and horizontal fairness. Finally, a relief supplies distribution model is built for the EMA that considers the actions of the non‐governmental organization (NGO) and donors. Both the utility and the fairness are considered in the objective of the proposed model. In numerical experiments, the Wenchuan earthquake is conducted to illustrate the applicability of the model and provide implications for decision‐makers. The results show the benefits of considering both vertical fairness and horizontal fairness. Then the suitable threshold is given by the analysis of the time delay. Finally, managerial insights and recommendations for the EMA and the NGO derived from the numerical experiments are presented. The findings in this paper help improve the reliability of the relief supplies distribution system.

A multi-period optimal distribution model of emergency resources for responding to COVID-19 under uncertain conditions

Ideally, optimal emergency resource allocation would have been vital for effective relief work during the COVID-19 outbreak. However, the suddenness of the epidemic and uncertainty of its spread added some difficulties to distributing emergency resources. First, this study introduces triangular fuzzy numbers to describe the uncertainty of supply and demand of emergency resources, and interval numbers to describe the time required for resource transportation under disaster conditions. To minimize the total delivery time and difference in the total satisfaction rate, this study constructs an optimal model for emergency resource distribution under uncertain conditions that considers both efficiency and equity. Subsequently, an improved genetic algorithm (IMGA) is proposed to obtain the optimal decision scheme. Finally, a case study on emergency resource distribution during the COVID-19 pandemic is conducted for model verification. The results demonstrate that the proposed model can improve the efficiency and effect of emergency resource distribution. The model allocates some emergency resources to each demand site during each emergency period, which can help avoid large losses caused by extreme shortages of resources at a certain demand point. The emergency resource allocation scheme considers the transportation time and degree of impact, which is beneficial for enhancing the flexibility of decision-making and practical applicability of distribution operations. A comparative analysis of the algorithms shows that the proposed IMGA is an effective method for managing emergency resource distribution optimization problems because it has higher solving efficiency, better convergence, and stronger stability. These findings can provide decision support for the optimal distribution of large-scale, multiperiod emergency resources during the COVID-19 pandemic.

Mathematical modeling of minimization of electricity consumption by industrial enterprises with continuous production

Relevance. Determined by the importance of minimizing electrical power consumption in industrial enterprises with continuous production, considering the specific characteristics of their technological processes and the requirements to maintain the output volume of their products. Aim. To solve the task of minimizing electrical power consumption based on a mathematical model and gradient method under optimal planning of the production volume of the industrial enterprises with continuous production; to develop a mathematical model for optimal distribution of production over a time cycle (month, quarter, year) across departments, taking into account both simple and functional constraints, derived from the condition of ensuring minimal electrical power consumption in industrial enterprises with continuous production. Methods. When developing the mathematical model for ensuring minimal electrical power consumption while preserving the production volume, classic Lagrange optimization methods were used. To ensure sufficient calculation accuracy, iterative methods were also applied. For the task under consideration, a calculation error margin of ε=0,1 was assumed and established. It is known that the choice of calculation error margin depends on the specifics of the problem at hand and the decision-maker. To verify the adequacy of the developed model, the method of finding the relative extremum of a function of several variables was used. Results. The use of the mathematical model, which takes into account the nature of the technological process and boundary conditions in both simple and integral forms, demonstrated the feasibility of optimal planning of electrical power consumption by the enterprise. The effectiveness of the developed approaches was verified using a metallurgical enterprise as an example of an industrial enterprise with continuous production, in solving the task of minimizing electrical power consumption for products produced during the reporting period. The use of the proposed model allowed for a reduction in annual electrical power consumption by 2.5% while maintaining the same production volume. One of the classic optimization methods – the method of finding the relative extremum of functions of several variables – showed almost identical results upon verification. This serves as further evidence of the adequacy of the proposed model.

A cold chain logistics distribution optimization model: Beijing-Tianjin-Hebei region low-carbon site selection

PurposeThis study reduces carbon emission in logistics distribution to realize the low-carbon site optimization for a cold chain logistics distribution center problem.Design/methodology/approachThis study involves cooling, commodity damage and carbon emissions and establishes the site selection model of low-carbon cold chain logistics distribution center aiming at minimizing total cost, and grey wolf optimization algorithm is used to improve the artificial fish swarm algorithm to solve a cold chain logistics distribution center problem.FindingsThe optimization results and stability of the improved algorithm are significantly improved and compared with other intelligent algorithms. The result is confirmed to use the Beijing-Tianjin-Hebei region site selection. This study reduces composite cost of cold chain logistics and reduces damage to environment to provide a new idea for developing cold chain logistics.Originality/valueThis study contributes to propose an optimization model of low-carbon cold chain logistics site by considering various factors affecting cold chain products and converting carbon emissions into costs. Prior studies are lacking to take carbon emissions into account in the logistics process. The main trend of current economic development is low-carbon and the logistics distribution is an energy consumption and high carbon emissions.