Research On Hazmat Transportation And Distribution Scheme Based On Dynamic Spatiotemporal Service Network

Hazardous materials(hazmat)transportation is related to the national economy and the people's livelihood.With the growth of China's industrialization,the demand for hazmat transportation has increased steadily in recent years.Once a transportation accident occurs,however,it will not only cause casualties and property losses,but also lead to pollution of air,soil,water,and environment,which has a great negative social impact.Therefore,it is quite important to reduce the threat to vehicles,personnel and facilities in the transportation network and to reduce the transportation risk when determining the hazmat transportation and distribution scheme.According to the distance and transportation zone,the hazmat transportation can be divided into long-distance inter-city rail-road transportation and short-distance inner-city distribution.The hazmat distribution scheme has obvious characteristics of both time and space.When determining a reasonable transportation route,the time-varying characteristics of each indicator of the route must be taken into account.The total hazmat transporting risk value can be reduced by optimizing the distribution scheme by avoiding high-risk periods.Two kinds of distribution schemes,which are inter-city hazmat transport by rail-road intermodal network and the inner-city road network,are studied in this paper,respectively.(1)According to the characteristics of hazmat transportation and the time-varying network,the time-space service network of rail-road multimodal transportation is constructed by increasing time dimension.While for the short-distance hazmat distribution scheme inner-city,the spatiotemporal service network which completes the distribution within the same period and the virtual spatiotemporal service network with more periods are considered respectively,so that the difference of time-varying hazmat distribution schemes between inter-city and inner-city is depicted accurately.(2)Based on the time-varying weight of transportation arcs,the multi-objective optimization model of single vehicle route for hazmat transportation is established in the spatiotemporal rail-road multimodal network,and is transformed into a single objective 0-1integer programming model with the minimum risk as the objective by using ?-constraint method.Additionally,a heuristic algorithm for the optimal route of the spatiotemporal network is designed based on Dijkstra labeling method.(3)For the case that multiple vehicles of hazmat transportation start continuously in the rail-road intermodal network,a multi-objective optimization model with the minimal total risk value,total transportation cost and total running time is built,combining with the constraints of the starting vehicles' departure time window,the minimum departure interval between adjacent vehicles and the longest duration of the convoy from the departure to the destination.Then the model is transformed into a single objective one,and an improved genetic algorithm embedded with Dijkstra method is designed.(4)In the inner-city road network,the hazmat distribution station node is introduced and time-varying weights network arcs are used.Moreover,the impedance(safety protection cost)functions of period-transport arcs are established,and a bi-level programming model based on user equilibrium theory is built to study the optimal hazmat distribution scheme for multiple users.The upper layer model ensures the total transportation cost of the distribution scheme to be minimum and the lower one promises the even and minimal hazmat flow impedance of each time-varying route.A improved genetic algorithm embedded with Frank-Wolfe is developed to solve the model.(5)A transfer node with the function of temporary parking of hazmat vehicles is added to the spatiotemporal service network.Meanwhile,a virtual waiting arc is introduced to represent their corresponding impedance value(safety protection cost)and the dwelling cost.Then,a bilevel programming model is established to solve the distribution scheme with the minimum transportation cost combining with the even hazmat flow distribution scheme,and an enhanced genetic algorithm is finally designed.Time-dimension is introduced in this paper to present the time-varying characteristic of determining hazmat transportation route and dispatching scheme.Then two hazmat spatiotemporal service networks,that is,a hazmat rail-road intermodal transportation and an inner-city transportation,are designed,which can consider the time fluctuation of multiple weights of hazmat transporting and dispatching scheme.Consequently,the shortest route of hazmat rail-road intermodal transportation and the optimal inner-city dispatching scheme can decrease the overall risk degree simultaneously.This research can provide the theoretical support to decision-makers to guarantee hazmat transportation security and to reduce threat to the transportation network.