WO2022266827A1

WO2022266827A1 - Logistics network simulation method and system

Info

Publication number: WO2022266827A1
Application number: PCT/CN2021/101406
Authority: WO
Inventors: Sheng Liu; Shengnan Wu; Liang Yan; Yu Wang; Jianwei Lin
Original assignee: Beijing Jingdong Zhenshi Information Technology Co., Ltd.
Priority date: 2021-06-22
Filing date: 2021-06-22
Publication date: 2022-12-29
Also published as: CN117730311A

Abstract

The present disclosure relates to logistics network simulation method and system. There provides an electronic device for logistics network simulation, the electronic device comprise a processing circuit configured to: acquire an event sequence including discrete logistic events that characterize the whole logistic process of a logistics network, and sequentially process logistic events in the event sequence to simulate the operation of the logistics network.

Description

Logistics Network Simulation Method and System

Field of the Invention

The present invention generally relates to logistics network, and particularly relates to simulation and construction of logistics network.

Background

With the progress of electronic commerce, the express logistics industry has gotten rapid development in recent years, a large number of logistics clusters have been built and operated all over the world, and up to now, about over ten million packages are delivered per day in China. A logistics enterprise manages a large-scale logistics network that contains hundreds of warehouses and sorting hubs, thousands or tens thousands of express stations, thousands of backbone transporting routes and tens of thousands of station transporting routes, and provides end-to-end logistics and express services which bring great convenience to people.

In recent years, the sales volume of grand promotion such as 11.11 and 6.18 of e-commerce websites such as Jindong and Taobao has reached new highs, which has brought great challenges to express industry, such as warehouse overloading, serious delay and package loss. In order to deal with these problems, express delivery enterprises need to appropriately plan the logistics network according to the hundreds of millions of package distribution needs during the promotion period. However, in the prior art, the traditional mathematical planning tools are difficult to solve such a large-scale problem, the planners lack effective tools, and mainly rely on experience to plan the logistics network, which is time-consuming, laborious, and with poor effect.

Therefore, it is still required to provide an improved solution for planning the logistics network optimally so as to effectively solve the problems and challenges in the prior art.

Unless otherwise stated, it should not be assumed that any of the methods described in this section are prior art simply by being included in this section. Also, unless otherwise stated, issues recognized with respect to one or more methods should not be assumed as being recognized in any prior art on the basis of this section.

Disclosure of the invention

The present disclosure proposes an improved mechanism for simulation of an actual logistics network, particularly based on logistic-related events during the logistic procedure throughout the actual logistics network.

The present disclosure further proposes an improved mechanism for optimally planning/scheduling an actual logistics network, and particularly, the planning/scheduling of the actual logistics network can be optimized based on the simulation. In this way, an actual logistics network, particularly a large-scale logistics network, can be optimized and thus can work smoothly and efficiently.

An aspect of the present disclosure relates to an electronic device for logistics network simulation, the electronic device comprise a processing circuit configured to acquire an event sequence including discrete logistic events which are arranged based on time series and characterize the whole logistic process of a logistics network, and sequentially process logistic events in the event sequence to simulate the operation of the logistics network.

Another aspect of the present disclosure relates to an electronic device for scheduling an actual logistics network, the electronic device comprises a processing unit configured to: acquire a package delivery demand intended to be handled by an actual logistics network; simulate package transportation process through the actual logistics network by the electronic device as described above, so as to obtain schedule arrangement for the actual logistics network based on the simulation result, and transmit information about the schedule arrangement for scheduling of the actual logistics network.

Yet another aspect of the present disclosure relates to a method of logistics network simulation, the method comprising steps of: acquiring an event sequence including discrete logistic events which are arranged based on time series and characterize the whole logistic process of a logistics network, and sequentially processing logistic events in the event sequence to simulate the operation of the logistics network.

Yet another aspect of the present disclosure relates to a method of scheduling an actual logistics network, the method comprising steps of: acquiring a package delivery demand intended to be handled by an actual logistics network; simulating package transportation process through the actual logistics network by the electronic device as described above, so as to obtain schedule arrangement for the actual logistics network based on the simulation result, and transmitting information about the schedule arrangement for scheduling of the actual logistics network.

Yet another aspect of the present disclosure relates to a device which includes a processor and a storage device, and the storage device stores executable instructions that, when executed by the processor, implement the method as described above.

Yet another aspect of the present disclosure relates to a non-transitory computer-readable storage medium storing executable instructions that, when executed by, for example, one or more processor, implement the method as described above.

Yet another aspect of the present disclosure relates to a computer program product which comprising instructions which, when executed by a computer, cause the computer to perform the method as previously described.

This section is provided to introduce some concepts in a simplified form that will be further described below in the detailed description. This section is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. Other aspects and advantages of the technology will become apparent from the following detailed description of the embodiments and the accompanying drawings.

Description of the drawings

The above and other objects and advantages of the present disclosure will be further described below with reference to specific embodiments and with reference to the drawings. In the drawings, the same or corresponding technical features or components will be denoted by the same or corresponding reference symbols.

Fig. 1 schematically illustrates a concept of logistics network simulation and planning/scheduling, according to an embodiment of the present disclosure.

Fig. 2 schematically illustrates a conceptual global package process in a large-scale logistics network, according to an embodiment of the present disclosure.

Fig. 3 schematically illustrates exemplary processes in a sorting hub including unloading, sorting, and loading process, according to an embodiment of the present disclosure.

Fig. 4 schematically illustrates exemplary sorting shift cycle and transporting cycle, according to an embodiment of the present disclosure.

Fig. 5A schematically illustrates an electronic device for simulation of a logistics network, according to an embodiment of the present disclosure, and Fig. 5B schematically illustrates a method for simulation of logistics network, according to an embodiment of the present disclosure.

Fig. 6 schematically illustrates exemplary event categories and events included in respective categories.

Fig. 7 schematically illustrates exemplary flowchart of simulation of a logistics network, according to an embodiment of the present disclosure.

Fig. 8A-8N schematically illustrates exemplary event processing logic according to an embodiment of the present disclosure respectively.

Fig. 9 schematically illustrates a logistics network covering area according to an embodiment of the present disclosure.

Fig. 10 schematically illustrates relation of events and procedures of the simulation engine for such events, according to an embodiment of the present disclosure.

Fig. 11 schematically illustrates an exemplary main procedure of the logistics network simulation engine, according to an embodiment of the present disclosure.

Fig. 12A schematically illustrates an electronic device for planning/scheduling an actual logistics network, according to an embodiment of the present disclosure, and Fig. 12B schematically illustrates a method for scheduling an actual logistics network by an artificial logistics network model, according to an embodiment of the present disclosure.

Fig. 13 schematically illustrates an exemplary framework of parallel logistics network mechanism, according to an embodiment of the present disclosure.

Fig. 14 schematically illustrates an exemplary flow of parallel logistics network mechanism, according to an embodiment of the present disclosure.

FIG. 15 illustrates an outline of a computer system in which embodiments according to the present disclosure may be implemented.

Fig. 16 illustrates the main interface of artificial logistics network and actual logistics network in the parallel logistics network system, according to an embodiment of the present disclosure.

The embodiments described in this section may be susceptible to various modifications and alternative forms, and specific embodiments thereof are shown by way of example in the drawings and are described in detail herein. It should be understood, however, that the drawings and detailed description thereof are not intended to limit the embodiments to the disclosed particular forms, but rather, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the claims.

Detailed description of preferred embodiments

Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the accompanying drawings. For clarity and conciseness, not all features of an embodiment are described in the specification. However, it should be understood that many implementation-specific settings must be made during the implementation of the embodiment in order to achieve specific goals of the developer, for example, to meet those restrictions related to equipment and business which may change depending on the implementation. In addition, it should also be understood that, although development work may be very complex and time-consuming, it is only a routine task for those skilled in the art benefiting from this disclosure.

Here, it should also be noted that, in order to avoid obscuring the present disclosure by unnecessary details, only processing steps and /or equipment structures that are closely related to at least the solution according to the present disclosure are shown in the drawings, while other details of little relevance to this disclosure are omitted.

In recent years, the express logistics industry has made rapid development, which has brought great convenience to people's lives. However, it brought some problems, such as long delivery time and high transportation cost for packages. Furthermore, for a grand promotion or an emergency scenario, there also exists some further problems, such as, sorting center overloading, too long delivery time of packages, even loss of packages, and so on.

Specifically, in one hand, during grand promotion period, such as 6.18 (June 18th) and 11.11 (November 11th) , because the number of orders from customers increased significantly, there may occur a serious backlog of packages in the logistics network. As a result, some nodes are shorthanded while some nodes are overstaffed. Some transporter routes still lacked vehicles while some transporter routes occupied too many vehicles. It not only increased transportation costs, but also leaded to more exhaust emissions. The whole logistics network is difficult to run smoothly and efficiently. On the other hand, the express logistics industry may confront some emergencies, for example, COVID-19, and in such a case, the logistics industry suffers challenges such as hub interruption, and cannot run smoothly and efficiently.

Usually, in a large scale logistics network, hundreds of warehouses and sorting hubs, thousands or tens of thousands of express stations are interconnected by tens of thousands of transporting routes, hundreds of thousands of people work together, hundreds of thousands of trucks, trains and planes run on transporting routes. Such a large-scale logistics network is obviously a typical complex system. Traditional approaches have great difficulty in predicting its load status and bringing out efficient resource allocation scheme so as to schedule or plan the logistics network. In particular, it is difficult to model and optimize the logistics network by traditional methods.

In current, various simulation relevant methods have been provided for studying complex systems, and at present, the simulation of logistics industry mainly focuses on the internal simulation of sorting center and warehouse, and the simulation of local process during package transportation, such as package station collection and package station dispatch, however, it is difficult to realize the whole network simulation of large-scale logistics network because of the requirements of large amount of data, large amount of computation and fast running, and currently, no researches are dealing with large scale logistics network simulation for a large-scale logistics network.

The present disclosure is proposed in view of the above, and proposes an improved concept of simulating and thereby planning/scheduling a logistics network, particularly a large-scale logistics network.

In one aspect, there proposes an improved simulation for a logistics network, wherein the logistics network simulation can be used for describing a whole process of all transported objects (packages) from a departure site to a destination site in a logistics network in a certain period of time, and particularly verify the logistics network performance under a certain logistics network configuration. The simulation of the present disclosure can be particularly applied to a whole logistics network, which generally refers to a large-scale logistics network of a certain logistics enterprise in the whole country or even in the whole world.

According to embodiments of the present disclosure, the logistics network simulation is based on appropriate logistic events which can characterize the whole logistic process throughout the logistics network, and thus a scheme or configuration can be efficiently and accurately simulated and verified for the actual logistics network. In particular, the logistics network simulation of the present disclosure is based on a discrete event simulation method, wherein the complex dynamic processes of logistics network can be abstracted into discrete events of certain categories, and such events can be processed sequentially in the order of event occurrence time to simulate the operation of logistics network, thus simplifying the description of logistics network and realizing efficient simulation of logistics network. For example, the simulation engine according to the embodiments of the present disclosure ensures that the simulation of 400 million packages distribution in 30 days is completed in half an hour on a personal computer, which significantly improved efficiency.

In another aspect, the present disclosure proposes a solution of scheduling/planning a logistics network based on the simulation of logistics network, wherein the configuration for the actual logistics network can be optimized based on the simulation, so that a large-scale logistics network to work smoothly and efficiently and achieve improved package delivery.

According to embodiments of the present disclosure, an optimized scheme may include optimal configuration parameters for the actual logistics network, and particularly, configuration parameters for various elements included in the actual logistics network. For example, the elements may include, for example, warehouse, sorting hub, expression station, transport route, vehicle, courier, and other appropriate elements in the logistics network. In operation, for each kind of elements, its configuration information, such as number, distribution including location information, resource allocation, operation mode including operation time, and any other information related to the functionality of the elements can be optimized by means of the simulation and applied to plan/schedule the kind of elements in the actual logistics network correspondingly, so that the actual logistics network can be optimized for a particular package delivery demand.

Hereinafter the embodiments of the present disclosure will be described with reference to the figures. Fig. 1 schematically illustrates the general basic concept of the present disclosure, which relates to logistics network simulation and logistics network planning/scheduling based thereon, and wherein the logistics network simulation interacts with the logistics network optimization.

In particular, the logistics network simulation is performed based on events characterizing the logistic transporting processes throughout the network, and the simulation can verify the logistics network performance by processing the events under a predetermined configuration information of the logistics network, such as that mentioned above. Such simulation and/or interaction can execute iteratively, until a desired logistics network metric is met or cannot got better, or a predetermined number of iterations have been reached. For example, the simulation can verify whether the predetermined configuration information can cause the logistics network to satisfy a desired metric and if not, the predetermined configuration information can be adjusted and then the simulation will be performed based on the adjusted configuration information. And when the simulation is complete, the corresponding configuration information can serve as optimized configuration information to plan/schedule the actual logistics network. During the simulation, the initial configuration information of the actual logistics network can be presented to the simulation side in any appropriate manner, for example, from the planning/scheduling side or from an appropriate third party.

Note that Fig. 1 is only illustrative and exemplary, instead of limiting its implementation manner. The structure as shown in Fig. 1 can be implemented in any appropriate manner, for example, the simulation side and planning/scheduling side can be implemented separately, or they can be integrated together, for example, implemented as a planning/scheduling entity for acquiring the logistics network information, performing simulation and then feedback the network configuration to the logistics network. Furthermore, the structure as shown in Fig. 1 can be incorporated into or can be outside of the logistics network.

Hereinafter, some embodiments of the present disclosure related to logistics network simulation will be described. According to embodiments of the present application, the simulation is performed to describe the package transportation processes in the logistics network. In particular, the packages will go through a variety of processes throughout the logistics network when transporting from a departure site to a destination site. Fig. 2 –Fig. 4 schematically illustrates three typical transportation relevant processes in large scale logistics network.

Fig. 2 describes the global package process in large scale logistics network. Packages are created by packing goods from warehouses or from customers. Then transporters carry them to sorting hubs where they are unloaded and sorted. After being sorted, they will be loaded to transporters and be carried to other sorting hubs or expression stations. In expression stations, packages are unloaded from transporters and delivered to customers. Packages can be carried from one sorting hub to another by road, rail, or air.

Fig. 3 illustrates the unloading, sorting, and loading process in a sorting hub. When a full truck arrives at a sorting hub, it can be unloaded only when the sorting hub has free unloading gates. Otherwise it should wait in a queue. When a truck has been unloaded, it releases an unloading gate and leaves from the sorting hub, and the earliest truck in the waiting queue will occupy the released unloading gate and be unloaded. Unloaded packages will be sorted immediately. After being sorted, packages will be allocated to a transporting route according to their destinations. After enough number of packages in a transporting route have been sorted, an empty truck will be called to load these packages if the sorting hub has free loading gates.

Fig. 4 (a) shows the sorting shift cycle in each sorting hub. A sorting hub usually works by two or three shifts. When a sorting shift in a sorting hub begins, packages will be unloaded from trucks which occupy unloading gates, and sorters will begin sort packages. If the sorting hub has free unloading gates and the waiting queue has full trucks, the full trucks will occupy the free unloading gates and be unloaded. When a sorting shift ends, packages pause being unloaded from trucks, and sorters pause sorting packages. Fig. 4 (b) shows the transporting cycle of each transporting route. A transporting route calls a transporter at a certain time before the departure deadline if the departure sorting hub of the transporting route has free unloading gates. The transporter begins loading sorted packages of the transporting route after it arrives. When the departure deadline of the transporting route is up, it will send all the transporters that belong to it.

Note that the processes as described above are only exemplary, and other logistic processes can take place during the package transportation, such as processes in other package processing location, processing stages, processing routes, and so on.

Fig. 5A illustrates an electronic device for simulating an actual logistics network. According to an embodiment, the electronic device 500 may comprise a processing circuit 502 configured to acquire an event sequence including discrete logistic events which characterize the whole logistic process of a logistics network, and sequentially process logistic events in the event sequence to simulate the operation of the logistics network.

In an embodiment, the simulation can be performed based on event classification/grouping, particularly resultant discrete logistics events, which are abstracted from a whole logistic process for package transportation throughout the logistics network from a departure site to a destination site in a certain period of time and which are representative. For example, the events categories may corrspondig to the processings in the logistics network which are dominant, or at the core, or are essential. In some embodiments, the logistics events can be classified in accordance with any appropriate conditions in the logistics network, for example, conditions about packages to be delivery, conditions about package processing, such as processing stages, processing locations such as sorting center, hub, etc., conditions about package delivery, such as delivery tools, delivery routes, and so on. Therefore, the present disclosure proposes a discrete event simulation technique, and it uses a small granularity where the loading (unloading, sorting, transporting) action of each package is described, so that the simulation can be performed accurately and efficiently.

In some embodiments, the logistic events to be carried out during logistic transportation can be classified into four categories, including sorting center category, route category, package category and transporter category, wherein the sorting center category encompasses some representative events occurring in the sorting center and also can be referred to as sorting category, the route category encompasses some representative vehicle scheduling events occurring on a route, the package category encompasses some representative events related to package scheduling, and the transporter category encompasses some representative events related to transporter scheduling.

Fig. 6 clusters exemplary logistic events which can be classified into four categories including 14 events. The four categories of events are sorting center events/sorting event category, route event category, package event category and transportation tools event category. Sorting center event category includes sorting shift start event (shift_begin) , sorting shift end event (shift_end) and periodic loading event (Periodic_loading_event) . Route event category includes route calling and loading event (route_call_empty_transporter) and route departure event (route_send_full_transporter) ; Package event category includes package initial arrival event (package_created) , package unloaded event (package_unloaded) , package sorted event (package_sorted) and package loaded event (package_loaded) . Transporter event category includes full transporter arrival event (full_transporter_arrived) , full transporter ready for unloading event (full_transporter_ready_unload) , empty transporter ready for loading event (empty_transporter_ready_load) , empty transporter departure event (empty_transporter_departed) and full transporter departure event (full_transporter_departed) . Such events are listed in Table 1.

Table 1. List of events that are handled by simulation engine

Note that such listed events are only exemplary, instead of being limitative. Alterantively or additionally, there may exist other event category, and each event category may include other events, depending on the logistic procedure in the logistics network as well as the adapted processing object and processing conditions. And any other appropriate event category and event numbers are encompassed in the present disclosure, as long as they can appropriately characterize the logistic procedure of packages in the logistics network, particularly the whole logistic process of packages in the logistics network.

In an embodiment of the present application, such events can be processed sequentially in any appropriate manner, for example, can be processed in order of event occurring time, or can be processed in order of priority, importance, weight, etc., so as to realize the operation of logistics network. In an embodiment, the events can be generated and added into an event sequence for simulation, and in the sequence, the events can be arranged and simulated in a variety of manners. In an example, the events can be arranged into an event sequence in time order and can be processed in a first-in first-out manner. That is, the events can be added into the sequence in the order of first to last generation, and the events which are first added into the sequence will be first simulated, that is, the earlier event will be first processed. In another example, the events in the event sequence is ordered and processed based on its weight or priority, which can be set for an event in consideration of the importance of the event, the priority of the packages related to the event, the rank of the packages related to the events, etc., and thus when simulated, an event with higher weight or priority will be first simulated.

Fig. 7 illustrates the flowchart of an exemplary logistics network simulation. At the start of simulation, the start time of simulation is set and an empty event sequence ES is generated, where any event, to be incorporated in the sequence ES, will be arranged in time order, and particularly arranged in ascending order of occurrence time, that is, the earliest event will be arranged first. Then, the logistic events starting from the simulation are generated and will be incorporated into the sequence. For example, the sorting center events occurring on the day the simulation starts are generated for each sorting center, including sorting shift start event and sorting shift end event for the first sorting shift, and the first cycle loading event, and the route events occurring on the day the simulation starts are generated for each route, such events can be incorporated into the sequence. Furthermore, although not shown, other kinds of events, such as package processing events, transporter events can be generated at the start of the simulation, such as such events occurring on the day the simulation starts, or can be generated along with logistic transportation during the simulation, and such events also shall be arranged into the sequence ES in occurring time order.

Next, the simulation is pushed forward by circularly fetching events from ES, particularly starting from the earliest event, advancing the simulation clock to the occurrence time of the earliest event, and processing the earliest event.

In particular, it will be judged whether there still exist any package in the logistics network which is not transported to its destination, and if so, an event occurring earliest will be fetched from the sequence ES and then be processed, while the simulation clock will be advanced to the occurrence time of the earliest event. Then the process will be executed in loop as long as there exist any package in the logistics network, which is not transported to its destination, and when all packages have been transported to their destination, the simulation can complete.

Hereinafter, the state of the actual logistics network corresponding to the occurrence time of each event and its processing logic will be described with reference to Figs. 8A-8N, which illustrates exemplary processes for respective events of respective categories as mentioned above.

Fig. 8A illustrates a processing logic for a sorting shift start event, wherein the occurrence time of the sorting center shift start event corresponds to the start state of a sorting shift of a sorting center in the actual logistics network.

Fig. 8B illustrates a processing logic for a sorting shift end event, wherein the occurrence time of sorting shift end event corresponds to the state of a sorting shift end instant of a sorting center in the actual logistics network.

Fig. 8C illustrates a processing logic for a periodic loading events, which means an event which is used to load package initial arrival events into the logistics network. Usually, the periodic loading event is generated with one day as the period and 00: 00: 00 every day as the occurrence time. The periodic loading event generates package initial arrival events in a sorting center within 24 hours after the occurrence time. In order to simplify the package arrival process without losing generality, it is assumed that all packages arrive at periodic discrete time points within 24 hours. For example, packages can be specified to arrive in batches every ten minutes. Accordingly, the processing logic for each periodic loading event will generate 144 package initial arrival events.

Fig. 8D illustrates a processing logic for a route calling and loading event, which corresponds to a state that in the actual logistics system, an empty vehicle is called for loading at the loading platform some time before the vehicle departing on the route, and the period of time of calling a vehicle in advance shall be appropriately set so as to ensure that the time the vehicle is waiting on the platform is not too long to waste loading and transportation resources, and is not too short to finish loading.

Fig. 8E illustrates a processing logic for a route departure event which corresponds to a state that a specific route in the actual logistics system reaches the departure time and a vehicle on the route is ready to start transportation from the starting point.

Fig. 8F illustrates a processing logic for a package initial arrival event which corresponds to a state that a batch of packages are collected for the first time and waiting for sorting at a certain moment in the actual logistics system.

Fig. 8G illustrates a processing logic for a package unloaded event which corresponds to a state that a package has just been unloaded from a transporter in the actual logistics system.

Fig. 8H illustrates a processing logic for a package sorted event which corresponds to a state that a package has just finished sorting in the actual logistics system.

Fig. 8I illustrates a processing logic for a package loading event which corresponds to a state that a package has just been loaded on a transporter in the actual logistics system.

Fig. 8J illustrates a processing logic for a full transporter arrival event which corresponds to a state that a transporter, such as vehicle, plane and other transportation tools, which is full of packages arrives at the destination sorting center and waits for unloading in the actual logistics system.

Fig. 8K illustrates a processing logic for a full transporter ready for unloading event which corresponds to a state that a transporter full of packages is ready to unload the first package in the actual logistics system.

Fig. 8L illustrates a processing logic for an empty transporter ready for loading event which corresponds to a state that an empty transporter is ready to be in place on the loading platform of the sorting center to load the first package in the actual logistics system.

Fig. 8M illustrates a processing logic for an empty transporter departure event which corresponds to an instant state when a transporter leaves the sorting center after total unloading in the actual logistics system.

Fig. 8N illustrates a processing logic for a full transporter departure event which corresponds to an instant state when a transporter leaves the sorting center after loading in the actual logistics system.

In some embodiments, additionally, the simulation can be performed based on the configuration/structure of the actual logistics network to simulate the package transportation throughout the actual logistics network. And such configuration information can be notified as simulation parameters. For example, the configuration can mean configuration about the whole of the logistics network, or can means configuration about a part of the logistics network, as along as such part can characterize the whole logistics network accurately. For example, the simulation can be performed based on an artificial logistics network which can cover a part or even the whole of the logistics network to imitate the actual logistics network, and can be implemented by software, hardware, firmware or any appropriate combination.

In an embodiment, the simulation can be performed in consideration of a package delivery demand, and particularly is performed to find optimized logistics network configuration which can meet the demand. The logistic configuration information can be edited, and particularly, can be dynamically established, edited, and updated based on the information of the actual logistics network, or can be dynamically updated iteratively during the simulation process. Such operation can be performed by the electronic device or even the processing circuit therein.

According to another embodiment, the simulation can be performed particularly under some constraints related to package transportation. For example, such constraints may include, not being limited to, package delivery demand, transportation entity constraint such as processing capacity limit, transportation time constraint, etc. Therefore, appropriate simulation can be performed to appropriately verify and obtain optimal scheme for scheduling the actual logistics network.

According to some embodiments, the simulation result can correspond to a certain logistics network configuration and may indicate the logistics network performance in the simulation period achieved by the configuration, so that it can determined whether the logistics network performance under such configuration meet a predetermined requirement, constraint, etc., and if not, the configuration of the logistics network can be modified and then the simulation will be re-performed based on the modified logistics network configuration. That is, the simulation is actually performed with respect to a predetermined logistics network configuration and aims to verify and finally obtain an optimized logistics network configuration, so that the actual logistics network can be optimized accordingly.

In another embodiment of the present disclosure, the simulation result may further include the simulated working status of the logistics network, for example, the working load of respective elements in the logistics network such as working load of warehouse, sorting hub, expression station, route, even vehicle in the route, which can be obtained and utilized to adjust and optimize the configuration of the actual logistics network. so that it can be judged whether the elements are overloaded or whether the capability of the elements have been fully utilized, and based on the judgement result, the capability of the elements can be appropriately adjusted. For example, if the elements are overload, the elements can be assigned more resources to enhance processing capability, and for example, if the capability of the elements is still redundant, some resources for the elements can be re-assigned to other elements.

In yet another embodiment, the simulation result may further include an information about package delivery cost, which can be obtained statistically with reference to the current configuration of the logistics network during simulation. In yet another embodiment, the simulation result may further include an information about logistics network load, which may indicate workload for elements in the network or whole workload of the network.

In an embodiment of the present disclosure, the simulation can be performed iteratively, and in an iteration, sequentially obtain events from an event sequence; and perform corresponding processing for the obtained events, and wherein when all events in the sequence have been performed, the operations in the round of iteration finish.

More specifically, in an embodiment, in each iteration, the simulation may include: simulating package delivery through the structure of the actual logistics network and previously given logistics network configuration; judge whether the simulation result satisfies a predetermined condition, if not, the logistics network configuration will be adjusted and serve as the basis for performing the next round of iteration, and else, the simulation terminates and the simulation result will be utilized as the final simulation result to derive the schedule arrangement, and in such a case, the schedule arrangement may correspond to the configuration parameters of the logistics network for obtaining the final simulation result.

According to an embodiment, the predetermined condition can indicate a predetermined package delivery requirement, which may be represented by a threshold related to the package delivery performance, such as a threshold for at least one of the package delivery duration, distance, sorting frequency, loss rate, etc., and the simulation result, that is, the simulated package delivery performance can be compared with the threshold, when the simulated package delivery performance exceeds the threshold, it means the current package delivery cannot satisfy the requirement, may cause poor user experience, and thus shall be further adjusted. For example, when the package delivery time is smaller than a specified time threshold and/or the package loss rate is smaller than a specified loss rate threshold, the simulation result can be deemed as satisfying the requirement.

According to another embodiment, the predetermined condition may indicate a predetermined number of rounds of iterations, and when the simulation has been performed the predetermined number of rounds, the simulation can terminate, and an optimal result can be selected from that of the predetermined number of rounds as the final simulation result.

According to an embodiment, the optimal result may correspond to a simulation result which is most cost-effective under constraints of the package delivery requirement, that is, the working cost for package delivery via the logistics network corresponding to the optimal result is lowest, and thus an optimal schedule arrangement for the actual logistics network can be obtained, which is cost optimal while the actual logistics network, when delivering the packages in accordance with the schedule arrangement, can satisfy the package delivery requirement. According to an embodiment, the optimal result may correspond to a simulation result which has appropriate workload under constraints of the package delivery requirement, for example, a score for each iteration can be generated to indicate the workload status and can be represented by a margin score, the larger the score is, the higher the network margin is and the reliability of the network is higher, and thus a simulation result with highest score can be selected. According to another embodiment, the optimal result may correspond to a simulation result which is most approximate to the package delivery requirement, can be selected, even all simulation results cannot satisfy the requirement. Here the most approximate means the difference between the package delivery result and the package delivery requirement is minimum. For example, the difference between simulated package delivery time and the required package delivery time is minimal, or the difference between simulated package loss rate and the required package loss rate is minimal.

According to another embodiments, the above two kinds of conditions can be combined, and in such as case, the simulation can be performed iteratively, wherein when the simulation result can satisfy the predetermined package delivery requirement, the simulation iteration can terminate, even the predetermined number of rounds is not reached. Otherwise, the simulation will be performed until the predetermined number of rounds is reached and an optimal result is selected.

According to the present disclosure, the processing circuit 502 can be implemented in a variety of manners, and particularly, may be in the form of a general-purpose processor, or may be a dedicated processing circuit, such as an ASIC. For example, the processing circuit 502 can be configured by a circuit (hardware) or a central processing device such as a central processing unit (CPU) . In addition, the processing circuit 502 may carry a program (software) for operating the circuit (hardware) or the central processing device. The program can be stored in a memory or an external storage medium connected from the outside, or downloaded via a network (such as the Internet) .

According to some embodiments, the processing circuit of the electronic device can include various units to implement the embodiments according to the present disclosure. For example, the processing circuit of the electronic device can include various units to implement various operations performed on the control device side described herein. In an embodiment, the processing circuit 502 may include an acquiring unit 504 for acquiring an event sequence including discrete logistic events which characterize the whole logistic process of a logistics network, and a simulation unit 506 for sequentially processing logistic events in the event sequence to simulate the operation of the logistics network. Additionally, when the simulation is performed iteratively, the simulation unit 506 can further include adjusting/updating unit for adjustment/updating of the simulation information for the next round of simulation, and judgement unit for judging whether the iteration terminates. Note that such units can be outside of the simulation unit 506. Although not shown, a communication unit may be included in the processing circuit 502, or even in the electronic device 500 but outside of the processing circuit 502, and is utilized to receive information as basis of simulation and transmit simulation result.

It should be noted that although the units are shown in the processing circuitry 502, it is only exemplary, and at least one of such units also can be outside of the processing circuit, even out of the electronic device. Each of the above units is only a logical module divided according to a specific function implemented by it, instead of being used to limit a specific implementation manner, and for example, such units as well as the processing circuit and even the electronic device may be implemented in software, hardware, or a combination of software and hardware. In an actual implementation, the foregoing units may be implemented as independent physical entities, or may be implemented by a single entity (for example, a processor (CPU or DSP, etc. ) , an integrated circuit, etc. ) . In addition, the above-mentioned respective units are shown with dashed lines in the drawings to indicate that these units may not actually exist, and the operations /functions they implement may be realized by the processing circuitry itself.

It should be understood that FIG. 5A is merely a schematic structural configuration of the electronic device, and optionally, the electronic device may further include other possible components not shown, such as a memory, a radio frequency stage for communication, a baseband processing unit, a network interface, a controller, and the like. The processing circuit may be associated with a memory and /or an antenna. For example, the processing circuitry may be directly or indirectly (e.g., other components may be connected in between) connected to the memory for data access. The memory may be a volatile memory and /or a non-volatile memory. For example, The memory 510 may include, but is not limited to, random access memory (RAM) , dynamic random-access memory (DRAM) , static random-access memory (SRAM) , read-only memory (ROM) , and flash memory, and the memory may store various kinds of information, for example, simulation information generated by the processing circuit 502, logistics network performance information, logistics network configuration parameters, package delivery demand information, etc., as programs and data for operation by the electronic device. The memory may also be located inside the electronic device for simulation but outside of the processing circuitry, or even outside of the electronic device for simulation.

According to an embodiment, there proposes a method of logistics network simulation, and Fig. 5B illustrates a flowchart of the method 600, including a step S602 of acquiring an event sequence including discrete logistic events which characterize the whole logistic process of a logistics network, and a step S604 of sequentially processing logistic events in the event sequence to simulate the operation of the logistics network. It should be noted that the method according to the present disclosure may further include operation steps corresponding to operations performed by the processing circuitry of the above-mentioned electronic device, which will not be described in detail here. It should be noted that each operation of the method according to the present disclosure may be performed by the aforementioned electronic device, in particular by a processing circuit or a corresponding unit, which will not be described in detail here.

Hereinafter, an exemplary simulation operation will be described based on a skeleton network of the whole logistics network which consists of nearly 1000 sorting hubs and thousands of transporting routes as shown in Fig. 9. Note that this is only an example, and the simulation can be implemented on more or less elements of the actual logistics network.

In an embodiment, the simulation model for the artificial logistics network can be appropriately created based on some information of the infrastructure of the actual logistics network as well as some constraints. Some necessary definitions for the logistics network model are listed in Table 2-4, which may correspond to the configuration information about the logistics network, and Table 5 lists the package transportation requirement, which may correspond to the package delivery demand. Note that such tables are only exemplary, and there may exist other appropriate tables defining other logistics structure or participant similarly, such as depending on their operation characteristics.

Table 2. The definition of sorting hub

Table 3. the definition of route

Table 4. the relation of sorting hub and route

Table 5. The package transportation requirement

And some constraints for the logistics network can be as follows. Note that such constraints are only exemplary, and other constraints can be set in accordance with the actual requirement of the logistics network.

A route can only appoint one type of default vehicle. Then

A route has only one origin and one destination. And the origin of a route cannot be same as its destination. So,

One route may have the same origin and the same destination with another route. Then,

The total volume of all packages in a vehicle cannot exceed the maximum allowed volume of the vehicle. Let Q denote the number of packages in a vehicle of the kth type. Let pack_vol _q (q=1, 2, …, Q) denote the volume of the qth package in the vehicle.

Thereby, a model for an actual logistics network, can be appropriately created, and based on such model, the simulation of the actual logistics network can be performed.

Fig. 10 schematically illustrates relation of events and procedures of the simulation engine for such events, according to an embodiment of the present disclosure, wherein each event will be triggered and handled by corresponding procedure invocation, and an exemplary processing flow of the logistics network simulation scheme LogisticsNetworkSimulation () according to the present disclosure will be described with reference to Fig. 11 which illustrates a main flow of parallel logistics network.

As parts of Experiment, experiment design module generates multiple schemes while experiment evaluation module evaluates schemes and picks out better ones by simulated network loads. A scheme consists of transporter arrangements of all transporting routes, staff arrangements in all sorting hubs etc. Simulated network loads are obtained by simulation engine by running the schemes. Usually billions of events are created and handled during the running of a large-scale logistics network simulation instance during grand promotion. In order to accelerate the simulation, some technologies such as asynchronous simulation, parallel simulation and distributed simulation can be applied to developing simulation engine.

When a simulation instance is started, a set of initial events are created and inserted into event queue. The types of initial events include shift_begin, line_call_empty_transporter and package_created. The simulation engine circularly gets out earliest event from event queue and deals with it by calling the HandleEvent () as shown in Fig. 10 until event queue does not contain any events. The input parameters of LogisticsNetworkSimulation () are defined in Table 1-Table 5, and the events and their corresponding handling can be implemented as described with reference to Figs. 8A-8N, and will be not described in detail here.

According to an embodiment of the present disclosure, the simulation result may include information about package delivery, and particularly, the information may package delivery duration, distance, sorting frequency, package delivery loss rate, etc. As an example, the simulation result consists of the values that are defined in Formula (7) - (9) .

Let PACK (N) = {p ₁, p ₂, …, p _N} denote a package set where p _n is the nth package in the set. Let created_time _n和delivered_time _ndenote the created time of p _n in its origin and the time when p _n is sorted in its destination, respectively. The average transition time of PACK (N) is expressed as:

Note that the delivery duration, that is, the transition time, can be calculated by the distance dividing the speed of the transporter. as shown in the formula (7) . Note that additionally, the transition time still can include some redundant time, including, not limited to, the waiting time during sorting, uploading, loading, and further some redundant time, and such redundant time can be set by default or by experience. Usually, the shorter the transition time is, the better the package delivery performance.

Let shipping_distance _n denote the shipping distance of p _n. The average shipping distance of PACK (N) is expressed as:

Let sorting_frequency _n denote the sorting frequency of p _n from its origin to its destination. The average sorting frequency of PACK (N) is expressed as:

Usually, the shorter the distance or sorting frequency is, the better the package delivery performance.

Hereinafter, the planning/scheduling of the logistics network according to the embodiments of the present disclosure will be described. In the present disclosure, a logistics network, particularly a large-scale logistics network, can be planned/scheduled based on the simulation as mentioned above, and particularly the simulation can verify and optimize the configuration for the logistics network and thereby the logistics network can be planed/scheduled based on the simulation.

Fig. 12A illustrates an electronic device for scheduling an actual logistics network, according to some embodiments of the present disclosure. According to an embodiment, the electronic device 1200 may comprise a processing circuit 1202 configured to acquire a package delivery demand intended to be handled by an actual logistics network; simulate package transportation process through the actual logistics network, so as to obtain schedule arrangement for the actual logistics network based on the simulation result, and schedule actual logistics network based on the schedule arrangement. Note that the simulation can be performed as mentioned above, so that an optimal planning/scheduling scheme can be obtained for the logistics network. Otherwise, the simulation can be performed in any other appropriate manner, as long as the simulation result can be appropriately communicated and utilized for the planning/scheduling of logistics network.

Hereinafter, the details of the processing according to some embodiments of the present disclosures will be described.

In the present disclosure, such scheme of planning/scheduling can be implemented by means of a parallel logistics network mechanism in which an artificial logistics network and an actual logistics network can cooperate with each other so as to improve the planning/scheduling of the actual logistics network. Specifically, an artificial logistics network can be established corresponding to the actual logistics network, and the simulation can be performed based on the artificial logistics network, so as to verify and optimize schemes for planning/scheduling of the actual logistics network, and in turn, the actual logistics network can be optimized based on the simulation result. The parallel logistics network greatly improves the ability of the logistics network to cope with dramatic increasement of package delivery amount and/or various emergency cases, such as COVID-19 and grand promotion.

Fig. 13 illustrates an exemplary parallel logistics network mechanism according to the present disclosure, and as illustrated in Fig. 13, the framework can include actual logistics network, artificial logistics network, logistics data extracting module and logistics scheme transmission module.

The actual logistics network can include/provide information about logistics infrastructures, logistics participants, and packages delivery demands. The logistics infrastructures may indicate various elements/nodes constituting the actual logistics network, such as sorting hubs, warehouses, routes, etc., as mentioned above. The logistics participants may indicate, for example, couriers, sorters, packagers, drivers, transporters, packaging machines, sorting machines and other persons or things that provide service for transporting the packages, which can also be deemed as elements of the actual logistics network. In a sense, the logistic infrasturcure and participants may belong to the configuration of the logistics network as mentioned above.

The package delivery demands may indicate the information about the package delivery throughout the logistics network, including the numbers of packages, requirement for package transportation, and so on. The package delivery demands data can be derived from historic package delivery demands and real time package delivery demands. In an embodiment, the package delivery demand can be a package delivery demand intended to be performed by the actual logistics network, and thus can be predicted from a current package delivery demand or a historical package delivery demand. Such prediction can be performed by any appropriate party in the system, and particularly can be performed by the planning/scheduling electronic device per se.In such a case, the electronic device may include a learning engine/unit which predicts future package delivery demands by current and historic package delivery demands data. In another example, the package delivery demands can be acquired from other entity, such as a third party which predicts the package delivery demands and present them to the electronic device. From this point, the electronic device can include an acquiring unit for acquiring the package delivery demands.

Note that in some sense, the actual logistics network as shown in Fig. 13 actually may mean at least one entity which can obtain and provide such information about the network structure and which can be implemented in any appropriate manner, such as a database, a computer/processor/controller for managing the actual logistics network.

The artificial logistics network actually means a virtual logistics network which can be established based on the information about the actual logistics network structure so as to has equivalent network structure with the actual logistics network and can be used for describing and simulating the functionalities/operations of the actual logistics network mathematically. Note that the expression “artificial logistics network” is only illustrative and is utilized to facilitate understanding, and the artificial logistics network can be implemented in a variety of manners. For example, the artificial logistics network can be implemented as an apparatus/device which can operate to simulate an actual logistics network, which can be implemented by, for example, hardware such as various chips, firmware, software such as that runs on computers. Particularly, the simulation device as mentioned above can implemented as a part of or the whole of the artificial logistics network as shown in Fig. 13, or even include the artificial logistics network.

The logistics data extracting module can obtain network structure and package data, such as the network configuration and package delivery demands, from the actual logistics network and transmits them to an artificial logistics network. Note that the logistics data extracting module can be implemented in a variety of manners, for example can be included in the actual logistics network or the artificial logistics network, for example, the above simulation device, as an communication unit, interface unit, and so on, or can be a third party forwarding the information about the actual logistics network and package delivery demand to the artificial logistics network.

The logistics scheme transmission module pushes the network load and treating schemes obtained by simulation/experiment at the artificial logistics network to the actual logistics network. Note that the logistics scheme transmission module can be implemented in a variety of manners, for example can be combined with the artificial logistics network, for example, the simulation device, as an communication unit, interface unit, and so on, or can be a third party forwarding the information about logistics scheme to the artificial logistics network.

Therefore, by means of the interaction between the artificial and actual logistics networks, the overall process continues running to drive optimal configuration for the actual logistics network so as to keep the actual logistics network running smoothly and efficiently. In particular, the simuliation will operate to verify and obtain a configuration which can satisfy the package delivery demand and/or achieve cost-efficient logistics network, and thus the actual logisitic network can be planned/scheduled based on the obtained configuration.

According to some embodiments of the present disclosure, the interaction between the artificial network and the actual network can be implemented in a variety of manners. According to an embodiment, the simulation can be performed iteratively and the artificial logistics network can be dynamically updated along with the iteration of simulation. Such updating can be performed in a variety of manners. In an example, when a round of iteration finishes, the parameters of the logistics network, particularly some configuration parameters for elements in the network may be appropriately adjusted as the basis for the next round of iteration, for example, an element whose workload is much higher or close to its working capability limit will be adjusted, such as shifting workload, enhance working capability, etc. In another example, the artificial logistics network can be even adjusted so as to add new elements or delete inappropriate elements, for example, warehouse, sorting hub, route or even vehicle can be newly added or deleted, and from this point, this kind of updating is especially available for establishment of an actual logistics network.

In an embodiment, the scheduling of the actual logistics network can be performed periodically or on demand. As an example, the scheduling operation can be performed at a predetermined time interval, so that the arrangement/configuration of the actual logistics network can be updated correspondingly and periodically. As another example, the scheduling operation can be performed in response to the operator’s request, triggered by a predetermined event, such as a grand promotion or an emergency. That is, when such event occurs, the scheduling operation can be performed accordingly.

In one embodiment, the electronic device may include a network constructor which creates artificial network structure based on logistics network data, and such network constructor can be even included in the processing circuit, or outside of the processing circuit. Such constructor may belong to a kind of modeling module for establishing a model for the actual logistics network.

According to the present disclosure, the processing circuit 1202 can be implemented in a variety of manners, similarly with that of the processing circuit 1202 and thus will not be described detailedly here. In an embodiment, the processing circuit 1202 may include an acquiring unit 1204 for acquiring a package delivery demand, a simulation unit 1206 for performing the simulation, and a scheduling unit 1208 for performing scheduling based on the schedule arrangement information. Of course, the processing circuit, even the electronic device can include other units, such as communication unit, memory, etc., as mentioned above and will not be described detailedly here.

According to an embodiment, there proposes a method of scheduling an actual logistics network by an artificial logistics network model, wherein the artificial logistics network model is created based on information about configuration of the actual logistics network, and Fig. 12B illustrates a flowchart of the method 1300. In step S1302, acquiring a package delivery demand intended to be handled by the actual logistics network; in step S1304, simulating package delivery through the actual logistics network based on the artificial logistics network model, so as to obtain schedule arrangement for the actual logistics network based on the simulation result, and in step S1306, scheduling the actual logistics network based on the schedule arrangement. It should be noted that the method may further include operation steps corresponding to operations performed by the processing circuitry of the above-mentioned electronic device for scheduling, which will not be described in detail here. It should be noted that each operation of the method according to the present disclosure may be performed by the aforementioned electronic device, in particular by a processing circuit or a corresponding unit, which will not be described in detail here.

Hereinafter, an exemplary processing flow of the parallel logistics network scheme according to the present disclosure will be described with reference to Fig. 14 which illustrates a main flow of parallel logistics network.

Network structure is created by network constructor according to network data. Future package delivery demands are obtained by leaning engine according to historic package delivery demands. Simulation model is created based on the network structure and future package delivery demands. Artificial logistics network invokes LogisticsNetworkSimulation () (defined in Procedure 1 as mentioned above) to run the logistics network simulation. As parts of Experiment, experiment design module generates multiple schemes while experiment evaluation module evaluates schemes and picks out better ones by simulated network loads. A scheme consists of transporter arrangements of all transporting routes, staff arrangements in all sorting hubs etc. Simulated network loads are obtained by simulation engine by running the model and schemes. Usually billions of events are created and handled during the running of a large-scale logistics network simulation instance during grand promotion. In order to accelerate the simulation, some technologies such as asynchronous simulation, parallel simulation and distributed simulation are applied to developing simulation engine. In an embodiment, the main procedure of the logistics network simulation engine can be performed as mentioned above, and will be omitted here.

It should be noted that the above description is only exemplary. The embodiments of the present disclosure can also be executed in any other appropriate manner, and the advantageous effects obtained by the embodiments of the present disclosure can still be achieved. Moreover, the embodiments of the present disclosure can also be applied to other similar application examples, and the advantageous effects obtained by the embodiments of the present disclosure can still be achieved. It should be understood that machine-executable instructions in the machine-readable storage medium or program product according to the embodiments of the present disclosure may be configured to perform operations corresponding to the above-mentioned device and method embodiments. When referring to the above embodiments of the device and method, the embodiments of the machine-readable storage medium or the program product are clear to those skilled in the art, and therefore will not be described repeatedly. Machine-readable storage media and program products for carrying or including the aforementioned machine-executable instructions also fall within the scope of the present disclosure. Such a storage medium may include, but is not limited to, a floppy disk, an optical disk, a magneto-optical disk, a memory card, a memory stick, and the like.

In addition, it should be understood that the series of processes and devices as described above may also be implemented by software and /or firmware. In the case of being implemented by software and /or firmware, a corresponding program constituting the corresponding software is stored in a storage medium of the related device, and when the program is executed, various functions can be achieved. As an example, a program constituting the software can be installed from a storage medium or a network to a computer having a dedicated hardware structure, such as a general-purpose computer 1100 shown in FIG. 15, and the computer is capable of executing various functions and so on when various programs are installed. FIG. 15 is a block diagram showing an exemplary structure of a computer as an example of an information processing apparatus that can be employed in an embodiment according to the present disclosure. In one example, the computer may correspond to the above-described exemplary electronic device on the intelligent service provider side or the electronic device on the application side according to the present disclosure.

In FIG. 15, a central processing unit (CPU) 1101 performs various processes according to a program stored in a read only memory (ROM) 1102 or a program loaded from a storage section 1108 to a random-access memory (RAM) 1103. In the RAM 1103, data required when the CPU 1101 executes various processes and the like is also stored as necessary.

The CPU 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input /output interface 1105 is also connected to the bus 1104.

The following components are connected to the input /output interface 1105: the input section 1106 including a keyboard, a mouse, etc.; the output section 1107 including a display, such as a cathode ray tube (CRT) , a liquid crystal display (LCD) , etc. and a speaker, etc.; the storage section 1108 including hard disks, etc.; and communication section 1109 including network interface cards such as LAN cards, modems, etc. The communication section 1109 performs communication processing via a network such as the Internet.

The driver 1110 can also be connected to the input /output interface 1105 as needed. The removable medium 1111 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is installed on the driver 1110 as needed, so that a computer program read out therefrom is installed into the storage section 1108 as needed.

In the case where the above-mentioned series of processing is realized by software, a program constituting the software is installed from a network such as the Internet or a storage medium such as a removable medium 1111.

Those skilled in the art should understand that a storage medium is not limited to the removable medium 1111 shown in FIG. 15 in which the program is stored, and which is distributed separately from the device to provide the program to the user. Examples of removable media 1111 include magnetic disks (including floppy disks) , optical disks (including CD-ROMs and digital versatile disks (DVDs) ) , magneto-optical disks (including mini disks (MD) (TM) ) and semiconductor memory. Alternatively, the storage medium may be ROM 1102, a hard disk included in the storage portion 1108, and the like, in which programs are stored, and are distributed to users along with the device containing them.

Exemplary applications

Hereinafter some exemplary applications in which the solution of the present disclosure can be employed will be described, wherein the applications are implemented by JD Logistics (JDL) which is a leading logistics company and the logistics arm of JD. com, the world’s third largest internet company by revenue, and in such applications, the solution of the present disclosure can effectively cope with the grand promotion as well as any possible emergency, and can improve configuration/planning of the actual logistics network so as to obtain advantageous result of package delivery..

By the above method and framework, parallel logistics network system can be implemented using Java programming language. The main interface of parallel logistics network system is illustrated in Fig. 16. Fig. 16 (a) is artificial logistics network sub-system while Fig. 16 (b) is actual logistics network sub-system.

In order to illustrate the accuracy of logistics network simulation engine, we firstly run the countermeasure obtained by parallel logistics network system on logistics network simulation engine. Then we compare the simulation result with the result of actual logistics network that are handled by the same countermeasure. The accordance rate of the static package routing of the two results is 99.7%while the accordance rate of the dynamic package routing of the two results is 94.2%. The latter is lower than the former because the countermeasure is slightly modified to dealing with unexpected situations (such as traffic accident, bad weather) during grand promotion. As a result, we can say parallel logistics network system is believable.

In one hand, the application of the present disclosure to a scenario similar to WUHAN-anti-COVID-19 period as an example of the emergency will be described, and in particular, the logistic data related to scenario similar to the WUHAN-anti-COVID-19 period can be obtained as exemplary data for evaluating the performance of the solution according to an embodiment of the present disclosure.

Specifically, during the WUHAN-anti-COVID-19 period in 2020, Wuhan which is the biggest city in Central China is locked down. Several sorting hubs that are located in Wuhan are closed. By means of the parallel logistics network, the countermeasures can be designed, verified, and optimized by artificial logistics network. And then the optimized countermeasure is forwarded to actual logistics network. Table 6 lists the average transition time and the average sorting frequency of packages in the affected ODs (Origin to Destination) by the lockdown of Wuhan. during the lockdown period of Wuhan, the average transition time of packages in the affected ODs (Origin to Destination) is 2.88 days while the average sorting frequency of packages in these ODs are 3.23. Before the lockdown period of Wuhan, the average transition time of packages in these ODs is 2.79 days while the average sorting frequency of packages in these ODs are 3.26. Parallel logistics Network minimize the impact of the lockdown of Wuhan during COVID-19 on the logistics network. The average transition time of packages in the affected ODs only increased by 3.2%.

Table 6. the comparison of average transition time and the average sorting frequency of packages

In another hand, the application of the present disclosure to the grand promotion period as an example of dramatic increasement of package orders will be described.

Specifically, the parallel logistics network can be utilized to optimize any scenario similar the JDL logistics network during the grand promotion period 6.18 (June 18th) and 11.11 (November 11th) . Artificial logistics network obtains logistics network structure and package delivery demands from actual logistics network. Then countermeasures are designed, verified, and optimized by artificial logistics network. And then the optimized countermeasure is forwarded to actual logistics network.

In the prior art, for such a grand promotion, traditional techniques may generate package delivery countermeasures for the grand promotion by experience. It took dozens of people a few days to formulate the countermeasure before each grand promotion period, and the working efficiency is low while the effectiveness of the countermeasure may be not enough. On the contrary, by means of the solution according the embodiment of the present disclosure, the countermeasure can be formulated efficiently and accurately, for example, one people is able to formulate the countermeasure in one day, while an improved package delivery result can be obtained.

As an example, the parallel logistics network system run logistics network based on the data from 2020/10/24 to 2020/11/24 during which the 11.11 grand promotion of JD is open and totally more than 400 million of packages are delivered. In order to illustrate the efficiency of parallel logistics network system for grand promotion, we firstly run the countermeasure for 11.11 grand promotion obtained by experience (CE for short) and the countermeasure 11.11 grand promotion obtained by parallel logistics network system (CP for short) on logistics network simulation engine, respectively. Here CP is obtained by 18 iterations of human-in-loop simulation and optimization on parallel logistics network system. It takes a PC (CPU: i7 8700HQ at 4.6GHz, 32GB memory) about less than half hour to run the one iteration of simulation of artificial logistics network. We classify all packages into three categories according to their delivery distances. The three categories are called short-distance (≤600 kilometers) , middle-distance (600 kilometers-1500 kilometers) and long distance (≥1500 kilometers) , respectively. In order to evaluate the simulation result in a steady status, we only analyze the simulation data from 2020/11/1 to 2020/11/15 rather than the one from 2020/10/24 to 2020/11/24. The average transition times of packages by CE and CP from 2020/11/1 to 2020/11/15 are listed in Table 7.

Table 7. The average transition times of packages by CE and CP from 2020/11/1 to 2020/11/15

From Table 7, the average transition time of short-distance packages, middle-distance packages, and long-distance packages from 2020/11/1 to 2020/11/15 by CP is 19.3%, 7.2%and 5.7%lower than the ones by CE, respectively.

The average shipping distances and sorting times of packages by CE and CP from 2020/11/1 to 2020/11/15 are listed in Table 8. The average shipping distance and sorting times of packages from 2020/11/1 to 2020/11/15 by CP is 6.6%and 2.1%lower than the ones by CE, respectively.

Table 8. The average shipping distances of packages by CE and CP from 2020/11/1 to 2020/11/15

By applying the countermeasure obtained by parallel logistics network system, the undeliverable packages are reduced significantly. The numbers of undeliverable packages by CE and CP from 2020/11/1 to 2020/11/15 are listed in Table 9. The average undeliverable packages by CP is 67.8%lower than the ones by CE.

Table 9. The number of undeliverable packages by CE and CP from 2020/11/1 to 2020/11/15

The parallel logistics network system is applied for the 6.18 and 11.11 grand promotions. The parallel logistics network system brings great convenience to the network planning department of JD Logistics. In a word, for a large logistics company, parallel logistics network system can improve its customer experience by reducing package transition time and cutting down undeliverable packages in its logistics network. Parallel logistics network system can also cut down transition cost and sorting cost by lessening package transition distance and package sorting times. Shorter package transition distance also means lower air pollution from trucks and other conveyance.

Although the present disclosure and its advantages have been described in detail, it should be understood that various changes, substitutions, and alterations can be made hereto without departing from the spirit and scope of the present disclosure as defined by the appended claims. Moreover, the terms "including" , "comprising" , or any other variation thereof, of the embodiments of the present disclosure are intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but includes other elements not explicitly listed, or also elements inherent to such a process, method, article, or device. Without more restrictions, the elements defined by the sentence "including a ... " do not exclude a case that in the process, method, article, or device that includes the elements, other identical elements exist.

Although some specific embodiments of the present disclosure have been described in detail, those skilled in the art should understand that the above embodiments are merely illustrative and do not limit the scope of the present disclosure. Those skilled in the art should understand that the above embodiments may be combined, modified, or replaced without departing from the scope and essence of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

An electronic device for logistics network simulation, the electronic device comprise a processing circuit configured to:

acquire an event sequence including discrete logistic events that characterize the whole logistic process of a logistics network, and

sequentially process logistic events in the event sequence to simulate the operation of the logistics network.
The electronic device of claim 1, wherein discrete logistic events are arranged in the order of event occurring time, and wherein the processing circuit is further configured to process the logistic events in the sequence in a manner of earliest-occurring-first-processing.
The electronic device of claim 1, wherein the events are ordered in the event sequence based on respective weights, wherein the weight is set for an event in consideration of at least one of the importance of the event, the priority of the package related to the event, the rank of the packages related to the events, and

wherein the processing circuit is further configured to process the logistic events in the sequence according to the weight from high to low.
The electronic device of claim 1, wherein the processing circuit is further configured to: perform corresponding processing for the events in the sequence, until all events in the sequence have been processed.
The electronic device of claim 1, wherein the simulation result is related to packages delivery performance including at least one of packing sorting frequency, package delivery duration, and package loss rate.
The electronic device of claim 1, wherein the processing circuit is further configured to process logistic events based on configuration of the logistics network, and the simulation result can be utilized to verify or modify the configuration of the logistics network for planning or scheduling the logistics network.
The electronic device of claim 1, wherein the processing circuit is further configured to:

perform the simulation of package delivery iteratively, and

when a predetermined condition is met, terminate the iteration, and determine a specific simulation result for deriving the schedule arrangement.
The electronic device of claim 7, wherein the predetermined condition indicates a package delivery requirement threshold, and wherein the processing circuit is further configured to:

perform the logistics network simulation iteratively, and

once the simulation result of a round of iteration is smaller than or equal to the package delivery requirement threshold, terminate the iteration and obtain the simulation result as the specific simulation result.
The electronic device of claim 7, wherein the predetermined condition indicates a predetermined number of rounds of iterations, and wherein the processing circuit is further configured to:

perform the simulation of package delivery through the predetermined number of rounds of iterations, and

select an optimal simulation result from the simulation results in the predetermined number of rounds of iterations as the specific simulation result.
The electronic device of claim 9, wherein the optimal simulation result is one of:

a simulation result which is most cost-effective while meeting the package delivery requirement;

a simulation result which has most optimal network workload while meeting the package delivery requirement; or

a simulation result which is most approximate to the package delivery requirement in a case the package delivery requirement is not met during the iteration.
An electronic device for scheduling an actual logistics network, the electronic device comprises a processing unit configured to:

acquire a package delivery demand intended to be handled by an actual logistics network;

simulate package transportation process through the actual logistics network by a simulation device of any of claims 1-10, so as to obtain schedule arrangement for the actual logistics network based on the simulation result, and

schedule the actual logistics network based on the information about the schedule arrangement.
The electronic device of claim 11, wherein the simulation is performed based on an artificial logistics network model, wherein the artificial logistics network model is created based on information about configuration of the actual logistics network.
The electronic device of claim 11, wherein the scheduling is performed periodically or on demand.
The electronic device of claim 11, wherein the package delivery demand is predicted based on a current package delivery demand or a historical package delivery demand for the actual logistics network.
The electronic device of claim 11, wherein the schedule arrangement for the actual logistics network correspond to configuration parameters of the artificial logistics network model when obtaining the simulation result.
The electronic device of claim 11, wherein the schedule arrangement for the actual logistics network include configuration information for respective elements included in the network comprising at least one of the number of elements, the distribution of elements, the working mode of elements, the working capability of elements.
A method for logistics network simulation, the method comprising:

acquiring an event sequence including discrete logistic events that characterize the whole logistic process of a logistics network, and

sequentially processing logistic events in the event sequence to simulate the operation of the logistics network.
A method for scheduling an actual logistics network, the method comprising:

acquiring a package delivery demand intended to be handled by an actual logistics network;

simulating package transportation process through the actual logistics network by a simulation device of any of claims 1-10, so as to obtain schedule arrangement for the actual logistics network based on the simulation result, and

scheduling the actual logistics network based on the information about the schedule arrangement.
A device comprising:

One or more processors; and

One or more storage media storing instructions that, when executed by the one or more processors, cause the method of any of claims 17-18 to be performed.
A computer-readable storage medium storing instructions that, when executed by one or more processors, cause a method of any of claims 17-18 to be performed.
A computer program product, comprising instructions which, when executed by a computer, cause the computer to perform the method of any of claims 17-18.