A pharmaceutical supply chain monitoring system and method based on artificial intelligence

Abstract

The present invention relates to the field of medicine supply, and specifically to an artificial intelligence-based medicine supply chain monitoring system and method, comprising: a supply scheduling module, a two-way delivery module, a cost management module, a path planning module and a redundancy utilization module. The supply scheduling module is used to record supply point and demand information, the two-way delivery module is used to select delivery points between supply points and demand points, the cost management module is used to determine the transportation cost function, the path planning module plans the transportation route, and the redundancy utilization module is used to store additional drugs at secondary supply points. The present invention can reduce the problem of insufficient distribution network coverage, while shortening the time it takes for drugs to reach patients from supply points, improving transportation timeliness, shortening transportation time, optimizing distribution routes, reducing drug transportation redundancy, improving the utilization rate of transportation resources, reducing drug supply chain costs, and improving the economic benefits of drug transportation.

Description

Technical Field

[0001] The present invention relates to the field of pharmaceutical supply, and in particular to an artificial intelligence-based pharmaceutical supply chain monitoring system and method. Background Art

[0002] The pharmaceutical supply chain encompasses the entire process of drug delivery, from raw material acquisition, manufacturing, distribution, and ultimately, patient delivery. It serves as the core support system for the healthcare industry. Drug supply is time-sensitive, temperature-sensitive, and subject to significant fluctuations in supply and demand. Therefore, the pharmaceutical supply chain requires highly responsive transportation to ensure timely delivery of drugs.

[0003] During the drug supply process, the road conditions from each delivery point to the supply point are complex, and the transportation network has a low coverage rate in remote areas, making it difficult to transport drugs to patients in a timely manner. Some supply points use the joint scheduling of enterprise ERP, logistics TMS and hospital HIS systems to enable vehicles to carry multiple drugs for continuous transportation at one time. However, the vehicle scheduling and route planning problems under high timeliness requirements are still difficult to solve, and the drug supply efficiency is low.

[0004] In addition, the cost of cold chain logistics is 2-3 times that of ordinary logistics. The transportation volume of medicines is small, the time is tight, and special vehicles are required. This causes the transportation vehicles to be semi-empty most of the time. Combined with the time delay costs and inventory costs under different distribution routes, the transportation costs are significantly increased. Complex scheduling plans are needed to achieve a balance between transportation safety, efficiency and cost. Summary of the Invention

[0005] The purpose of the present invention is to provide a pharmaceutical supply chain monitoring system and method based on artificial intelligence to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: an artificial intelligence-based pharmaceutical supply chain monitoring system, comprising: a supply scheduling module, a two-way delivery module, a cost management module, a path planning module, and a redundancy utilization module;

[0007] The supply scheduling module is used to record the location of supply points, drug inventory, cold chain capacity, and loading and unloading capabilities, and to receive real-time drug demand information. The drug demand information includes patient status, drug demand type, demand quantity, and demand location, and is marked on an electronic map to form a drug supply map. For each demand information, the drug shelf life and minimum transportation space are determined based on the drug type;

[0008] The bidirectional delivery module is used to preset transportation cost weights and transportation distance limits for patients based on demand information, select a delivery point between the patient and the supply point, and the delivery point is located on the transportation network so that the weighted sum of the distance from the transport vehicle to the delivery point and the distance from the patient to the delivery point is minimized. The vehicle waiting time is determined by the difference between the shelf life of the drug and the transportation time from the supply point to the delivery point. The number of dispatches is determined based on the drug transportation space demand, and the transportation tasks with the minimum transportation time between the delivery points being less than the difference in the shelf life of the drugs are merged, and a cost function between the transportation cost and the transportation distance of a single task is output.

[0009] The cost management module is used to plan the transportation routes from the starting supply point to the delivery point according to the road network topology, real-time road conditions, traffic restrictions and travel costs. Each transportation route corresponds to a transportation plan. The cost function is processed according to the actual distance and additional cost in the transportation plan to obtain the objective function of each transportation plan;

[0010] The path planning module is used to plan the objective function through a meta-heuristic algorithm to minimize the sum of the drug transportation cost and the patient transportation cost, while meeting the constraints of drug timeliness, vehicle capacity and transportation flow, output the route with the lowest transportation cost, use it as the planned route for vehicle supply, and determine the departure time of the vehicle and the patient and the estimated transportation time;

[0011] The redundancy utilization module is used to predict the use of medicines in a single area. When the vehicle is not fully loaded, additional medicines are stored in the remaining space. The types of stored medicines are the most frequently used medicines in the area where the last delivery point of the mission is located, and the shelf life is longer than the mission transportation time. The additional medicines are stored at the secondary supply point in the area. When the demand for medicines in the area is less than the demand for storing medicines at the secondary supply point, the secondary supply point will be used for transportation.

[0012] Furthermore, the supply scheduling module includes: a transportation network unit and a demand input unit;

[0013] The transportation network unit is used to input a map of the drug supply area through ArcGIS or QGIS map software to identify the road transportation network of the supply area;

[0014] The demand input unit is used to record the supply point location, drug inventory, cold chain capacity and loading and unloading capacity through dynamic warehousing and storage tools, and receive drug usage demand.

[0015] Furthermore, the two-way delivery module includes: an intermediate delivery unit, a dispatch planning unit, and a cost accounting unit;

[0016] The intermediate delivery unit is used to divide the service range of the supply point based on the Voronoi diagram and allocate patients to the nearest delivery point;

[0017] The dispatch planning unit is used to determine the number of vehicles to be dispatched and the dispatch distance, and to combine the transportation routes of different drugs using the Clarke-Wright economizing algorithm;

[0018] The cost accounting unit is used to initialize the distance cost of drug transportation according to the distance from the supply point to the delivery point to obtain a cost function.

[0019] Furthermore, the cost management module includes: a transportation route unit and a target cost unit;

[0020] The transport route unit is used to plan all available transport routes according to the road traffic status and keep the transport distance below a threshold;

[0021] The target cost unit is used to predict the loss cost and time cost of a vehicle passing through a road section, and accumulate them to obtain an objective function.

[0022] Furthermore, the path planning module includes: a solution iteration unit and a path selection unit;

[0023] The solution iteration unit is used to adopt an adaptive ALNS large neighborhood search dynamic adjustment strategy, update the traffic efficiency weight in real time, and plan the objective function;

[0024] The path selection unit is used to determine a route with the lowest transportation cost and meeting all constraints for transportation.

[0025] Furthermore, the redundancy utilization module includes: a demand forecasting unit, an additional supply unit, and a secondary transportation unit;

[0026] The demand forecasting unit is used to determine the mean and variance of drug usage frequency according to historical drug supply data and forecast the drug demand range;

[0027] The additional supply unit is used to introduce a packing problem algorithm to pre-allocate medicines to vehicles so that the space utilization rate of the vehicles reaches a threshold;

[0028] The secondary transport unit is used to establish a secondary supply system, and when the secondary supply point meets the demand, the secondary supply point will transport the medicine.

[0029] A pharmaceutical supply chain monitoring method based on artificial intelligence includes the following steps:

[0030] Step S1. Identify the road transportation network of the pharmaceutical supply area on the GIS map, enter the location of the pharmaceutical supply point and the pharmaceutical inventory, cold chain capacity, and loading and unloading capacity of the supply point, and receive and record pharmaceutical demand information;

[0031] Step S2. Set the patient's transportation cost weight and transportation distance limit based on the drug demand information, and select a delivery point between the patient and the supply point so that the delivery point is located on the transportation network and the weighted sum of the distance from the supply point to the delivery point and the distance from the patient to the delivery point is minimized;

[0032] Step S3. For each delivery point, a transport task is created from the supply point to the delivery point. The inter-delivery time between delivery points is calculated through road planning. Transport tasks with inter-delivery times shorter than the difference in the drug's shelf life are merged. Vehicles are arranged according to the transport task, and the transport cost is initialized.

[0033] Step S4. For each delivery task, a vehicle transport route is planned. The additional cost of each road segment is estimated based on the road network topology, real-time road conditions, traffic restrictions, and travel costs. The additional cost is added to the initial cost to obtain an objective function. This objective function is then planned to minimize transportation costs while meeting the timeliness of the drug, vehicle capacity, and transportation distance constraints. The transport route with the lowest transportation cost is determined.

[0034] Step S5. Determine the departure time and estimated transportation time of the vehicle and the patient. When the vehicle capacity is redundant, load the most frequently used drugs in the area where the last delivery point of the task is located in the redundant space and store the drugs at the secondary supply point in the area.

[0035] Furthermore, step S1 includes:

[0036] Step S11. Demarcate the service area of ​​a single supply point based on the Voronoi diagram. Use ArcGIS or QGIS mapping software to enter the regional transportation network within the supply point's service area. Mark the supply point location on the map. Dynamic inbound and outbound warehousing tools record and update the supply point's location, drug inventory, cold chain capacity, and loading and unloading capabilities.

[0037] Step S12: The cloud platform receives drug supply demands within the service scope and generates drug demand information, which includes patient status, drug demand type, demand quantity, and demand location.

[0038] Furthermore, step S2 includes:

[0039] Step S21. Set the patient's transportation cost weight according to the patient's status, satisfying Q1 = k1·Li, where Q1 is the cost of transporting the patient, k1 is the cost required to transport the patient per unit distance, and Li is the transportation distance from patient i to the delivery point;

[0040] Step S22. Determine the distance between the supply point and the demand point, calculate the weighted sum Y of the distance from the supply point to the delivery point and the distance from the patient to the delivery point, satisfying Y = k1·Li+k2·LR, where LR is the distance from the supply point to the delivery point, Li+LR is equal to the distance from the supply point to the patient, k2 is the basic cost required for the supply point to transport the drug one unit distance, and determine a delivery point on the transportation network to minimize the value of Y.

[0041] Furthermore, step S3 includes:

[0042] Step S31. Determine the delivery point for each piece of drug demand information within the cycle, and create a transportation task for each delivery point. The transportation task includes: the supply point location, the delivery point location, the type of drug to be transported, and the quantity of the drug;

[0043] Step S32: Plan the transportation routes between the delivery points and determine the connecting transportation time tr between the delivery points, satisfying tr = te + Lr / v, where Lr is the length of the transportation route between the delivery points, v is the vehicle transportation speed, and te is the time loss, which is determined by the congestion level of the transportation route and the delivery time of the drug;

[0044] Step S33. If the duration of the subsequent transportation between the delivery points is less than the difference in the shelf life of the drugs, the transportation tasks corresponding to the drugs are merged, and the Clarke-Wright saving algorithm is used to determine the transportation path of the vehicle after the merged transportation. For each transportation task, the vehicle transportation weight is multiplied by the total distance of the transportation task to obtain the initial transportation cost.

[0045] Furthermore, step S4 includes:

[0046] Step S41. Plan all transportation routes from the supply point to the delivery point with a transportation distance below a threshold. Determine the additional cost of each road segment based on the road network topology, real-time road conditions, traffic restrictions, and travel cost planning. The additional cost includes distance cost and travel cost. Add the additional cost to the initial cost to obtain the objective function.

[0047] Step S42: Plan the minimization objective function:

[0048]

[0049] Among them, Q is the objective function, Q0 is the initial transportation cost, k represents the section number of the transportation route, n is the number of sections, c k represents the additional cost of road segment k, t kis the additional time cost of road section k, including the time extension caused by extended distance, speed loss and poor road conditions, w is the delay cost per unit time, Σk1·Li represents the cost of transporting all patients to the delivery points, t0 is the estimated time for the vehicle to arrive at the first delivery point, m is the number of delivery points, tr i is the length of the subsequent transportation to the i-th delivery point, T m is the shelf life of the drug at the mth delivery point, ΣU m is the total volume of drugs, U0 is the vehicle loading capacity, Li is the transportation distance from patient i to the delivery point, L max The upper limit of the distance for transporting patients;

[0050] Step S43. All transportation routes are planned in succession, and the transportation route that meets the constraints of drug timeliness, vehicle capacity and transportation distance and has the lowest transportation cost is determined for drug transportation.

[0051] Furthermore, step S5 includes:

[0052] Step S51. Predict drug usage within the sub-regions of each delivery point. If the vehicle is not fully loaded, additional medication is stored in the remaining space. The medications stored are those with the highest frequency of use in the sub-region of the last delivery point of the mission and whose shelf life is greater than the mission transportation time.

[0053] Step S52: The extra drugs are stored at the secondary supply point of the sub-region. When the drug demand of the sub-region is less than the demand for the drugs stored at the secondary supply point, the secondary supply point is used to supply and transport the drugs.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. The present invention determines the maximum supply time and transportation space requirements of drugs based on drug types, selects a delivery point between the patient and the supply point, and sets a transportation cost weight and transportation distance limit for the patient. The vehicle waiting time is determined by the transportation time of the planned route, and the number of dispatches is determined by the drug transportation space requirements. The starting point and end point of the drug are transported in coordination, reducing the problem of insufficient distribution network coverage, shortening the time it takes for the drug to reach the patient from the supply point, and improving transportation timeliness.

[0056] 2. The present invention uses an iterative algorithm to plan the transportation efficiency and transportation cost of the transportation sections between the supply point and each delivery point, so as to minimize the sum of the drug transportation cost and the patient transportation cost and meet the maximum supply time limit, thereby determining the planned route. By adjusting the warehouse layout, transportation route and inventory level, the transportation time is shortened, the distribution path is optimized, and the redundancy of drug transportation is reduced.

[0057] 3. The present invention predicts the area where medicines are used, stores additional medicines in the remaining space of the vehicle, and stores them at secondary supply points in the area. When the demand for medicines in the area is less than the demand for storing medicines at the secondary supply points, the secondary supply points will transport the medicines, rationally utilizing redundant transportation capacity, improving the utilization rate of transportation resources, realizing the rapid deployment of commonly used medicines, reducing the cost of the medicine supply chain, and improving the economic benefits of medicine transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0059] Figure 1 This is a schematic diagram of the structure of an artificial intelligence-based pharmaceutical supply chain monitoring system of the present invention;

[0060] Figure 2 This is a schematic diagram of the steps of an artificial intelligence-based pharmaceutical supply chain monitoring method of the present invention. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0062] See also Figure 1 , the present invention provides a technical solution: an artificial intelligence-based pharmaceutical supply chain monitoring system, comprising: a supply scheduling module, a two-way delivery module, a cost management module, a path planning module and a redundancy utilization module;

[0063] The supply scheduling module is used to record the location of supply points, drug inventory, cold chain capacity, and loading and unloading capabilities, and to receive real-time drug demand information. The drug demand information includes patient status, drug demand type, demand quantity, and demand location, and is marked on an electronic map to form a drug supply map. For each demand information, the drug shelf life and minimum transportation space are determined based on the drug type;

[0064] The supply scheduling module includes: a transportation network unit and a demand input unit;

[0065] The transportation network unit is used to input a map of the drug supply area through ArcGIS or QGIS map software to identify the road transportation network of the supply area;

[0066] The demand input unit is used to record the supply point location, drug inventory, cold chain capacity and loading and unloading capacity through dynamic warehousing and storage tools, and receive drug usage demand.

[0067] The bidirectional delivery module is used to preset transportation cost weights and transportation distance limits for patients based on demand information, select a delivery point between the patient and the supply point, and the delivery point is located on the transportation network so that the weighted sum of the distance from the transport vehicle to the delivery point and the distance from the patient to the delivery point is minimized. The vehicle waiting time is determined by the difference between the shelf life of the drug and the transportation time from the supply point to the delivery point. The number of dispatches is determined based on the drug transportation space demand, and the transportation tasks with the minimum transportation time between the delivery points being less than the difference in the shelf life of the drugs are merged, and a cost function between the transportation cost and the transportation distance of a single task is output.

[0068] The two-way delivery module includes: an intermediate delivery unit, a dispatch planning unit and a cost accounting unit;

[0069] The intermediate delivery unit is used to divide the service range of the supply point based on the Voronoi diagram and allocate patients to the nearest delivery point;

[0070] The dispatch planning unit is used to determine the number of vehicles to be dispatched and the dispatch distance, and to combine the transportation routes of different drugs using the Clarke-Wright economizing algorithm;

[0071] The cost accounting unit is used to initialize the distance cost of drug transportation according to the distance from the supply point to the delivery point to obtain a cost function.

[0072] The cost management module is used to plan the transportation routes from the starting supply point to the delivery point according to the road network topology, real-time road conditions, traffic restrictions and travel costs. Each transportation route corresponds to a transportation plan. The cost function is processed according to the actual distance and additional cost in the transportation plan to obtain the objective function of each transportation plan;

[0073] The cost management module includes: a transportation route unit and a target cost unit;

[0074] The transport route unit is used to plan all available transport routes according to the road traffic status and keep the transport distance below a threshold;

[0075] The target cost unit is used to predict the loss cost and time cost of a vehicle passing through a road section, and accumulate them to obtain an objective function.

[0076] The path planning module is used to plan the objective function through a meta-heuristic algorithm to minimize the sum of the drug transportation cost and the patient transportation cost, while meeting the constraints of drug timeliness, vehicle capacity and transportation flow, output the route with the lowest transportation cost, use it as the planned route for vehicle supply, and determine the departure time of the vehicle and the patient and the estimated transportation time;

[0077] The path planning module includes: a solution iteration unit and a path selection unit;

[0078] The solution iteration unit is used to adopt an adaptive ALNS large neighborhood search dynamic adjustment strategy, update the traffic efficiency weight in real time, and plan the objective function;

[0079] The path selection unit is used to determine a route with the lowest transportation cost and meeting all constraints for transportation.

[0080] The redundancy utilization module is used to predict the use of medicines in a single area. When the vehicle is not fully loaded, additional medicines are stored in the remaining space. The types of stored medicines are the most frequently used medicines in the area where the last delivery point of the mission is located, and the shelf life is longer than the mission transportation time. The additional medicines are stored at the secondary supply point in the area. When the demand for medicines in the area is less than the demand for storing medicines at the secondary supply point, the secondary supply point will be used for transportation.

[0081] The redundancy utilization module includes: a demand forecasting unit, an additional supply unit and a secondary transportation unit;

[0082] The demand forecasting unit is used to determine the mean and variance of drug usage frequency according to historical drug supply data and forecast the drug demand range;

[0083] The additional supply unit is used to introduce a packing problem algorithm to pre-allocate medicines to vehicles so that the space utilization rate of the vehicles reaches a threshold;

[0084] The secondary transport unit is used to establish a secondary supply system, and when the secondary supply point meets the demand, the secondary supply point will transport the medicine.

[0085] like Figure 2 As shown, a pharmaceutical supply chain monitoring method based on artificial intelligence includes the following steps:

[0086] Step S1. Identify the road transportation network of the pharmaceutical supply area on the GIS map, enter the location of the pharmaceutical supply point and the pharmaceutical inventory, cold chain capacity, and loading and unloading capacity of the supply point, and receive and record pharmaceutical demand information;

[0087] Step S1 includes:

[0088] Step S11. Demarcate the service area of ​​a single supply point based on the Voronoi diagram. Use ArcGIS or QGIS mapping software to enter the regional transportation network within the supply point's service area. Mark the supply point location on the map. Dynamic inbound and outbound warehousing tools record and update the supply point's location, drug inventory, cold chain capacity, and loading and unloading capabilities.

[0089] Step S12: The cloud platform receives drug supply demands within the service scope and generates drug demand information, which includes patient status, drug demand type, demand quantity, and demand location.

[0090] Step S2. Set the patient's transportation cost weight and transportation distance limit based on the drug demand information, and select a delivery point between the patient and the supply point so that the delivery point is located on the transportation network and the weighted sum of the distance from the supply point to the delivery point and the distance from the patient to the delivery point is minimized;

[0091] Step S2 includes:

[0092] Step S21. Set the patient's transportation cost weight according to the patient's status, satisfying Q1 = k1·Li, where Q1 is the cost of transporting the patient, k1 represents the cost required to transport the patient per unit distance, and Li is the transportation distance from the patient to the delivery point;

[0093] Step S22. Determine the distance between the supply point and the demand point, calculate the weighted sum Y of the distance from the supply point to the delivery point and the distance from the patient to the delivery point, satisfying Y = k1·Li+k2·LR, where LR is the distance from the supply point to the delivery point, Li+LR is equal to the distance from the supply point to the patient, k2 is the basic cost required for the supply point to transport the drug one unit distance, and determine a delivery point on the transportation network to minimize the value of Y.

[0094] Step S3. For each delivery point, a transport task is created from the supply point to the delivery point. The inter-delivery time between delivery points is calculated through road planning. Transport tasks with inter-delivery times shorter than the difference in the drug's shelf life are merged. Vehicles are arranged according to the transport task, and the transport cost is initialized.

[0095] Step S3 includes:

[0096] Step S31. Determine the delivery point for each piece of drug demand information within the cycle, and create a transportation task for each delivery point. The transportation task includes: the supply point location, the delivery point location, the type of drug to be transported, and the quantity of the drug;

[0097] Step S32: Plan the transportation routes between the delivery points and determine the connecting transportation time tr between the delivery points, satisfying tr = te + Lr / v, where Lr is the length of the transportation route between the delivery points, v is the vehicle transportation speed, and te is the time loss, which is determined by the congestion level of the transportation route and the delivery time of the drug;

[0098] Step S33. If the duration of the subsequent transportation between the delivery points is less than the difference in the shelf life of the drugs, the transportation tasks corresponding to the drugs are merged, and the Clarke-Wright saving algorithm is used to determine the transportation path of the vehicle after the merged transportation. For each transportation task, the vehicle transportation weight is multiplied by the total distance of the transportation task to obtain the initial transportation cost.

[0099] Step S4. For each delivery task, a vehicle transport route is planned. The additional cost of each road segment is estimated based on the road network topology, real-time road conditions, traffic restrictions, and travel costs. The additional cost is added to the initial cost to obtain an objective function. This objective function is then planned to minimize transportation costs while meeting the timeliness of the drug, vehicle capacity, and transportation distance constraints. The transport route with the lowest transportation cost is determined.

[0100] Step S4 includes:

[0101] Step S41. Plan all transportation routes from the supply point to the delivery point with a transportation distance below a threshold. Determine the additional cost of each road segment based on the road network topology, real-time road conditions, traffic restrictions, and travel cost planning. The additional cost includes distance cost and travel cost. Add the additional cost to the initial cost to obtain the objective function.

[0102] Step S42: Plan the minimization objective function:

[0103]

[0104] Among them, Q is the objective function, Q0 is the initial transportation cost, k represents the section number of the transportation route, n is the number of sections, c k represents the additional cost of road segment k, t k is the additional time cost of road section k, including the time extension caused by extended distance, speed loss and poor road conditions, w is the delay cost per unit time, Σk1·Li represents the cost of transporting all patients to the delivery points, t0 is the estimated time for the vehicle to arrive at the first delivery point, m is the number of delivery points, tr i is the length of the subsequent transportation to the i-th delivery point, T m is the shelf life of the drug at the mth delivery point, ΣU m is the total volume of drugs, U0 is the vehicle loading capacity, Li is the transportation distance from patient i to the delivery point, L max The upper limit of the distance for transporting patients;

[0105] Step S43. All transportation routes are planned in succession, and the transportation route that meets the constraints of drug timeliness, vehicle capacity and transportation distance and has the lowest transportation cost is determined for drug transportation.

[0106] Step S5. Determine the departure time and estimated transportation time of the vehicle and the patient. When the vehicle capacity is redundant, load the most frequently used drugs in the area where the last delivery point of the task is located in the redundant space and store the drugs at the secondary supply point in the area.

[0107] Step S5 includes:

[0108] Step S51. Predict drug usage within the sub-regions of each delivery point. If the vehicle is not fully loaded, additional medication is stored in the remaining space. The medications stored are those with the highest frequency of use in the sub-region of the last delivery point of the mission and whose shelf life is greater than the mission transportation time.

[0109] Step S52: The extra drugs are stored at the secondary supply point of the sub-region. When the drug demand of the sub-region is less than the demand for the drugs stored at the secondary supply point, the secondary supply point is used to supply and transport the drugs.

[0110] Example: Three drugs are in demand during the same time period. Their shelf lives are 6 hours, 10 hours, and 12 hours, respectively. The shortest road network distances to the supply points are 120 km, 240 km, and 80 km, respectively. A delivery point is selected between the patient and the supply point. The patient transportation cost is three times the drug transportation cost. Therefore, delivery points at 90 km, 180 km, and 60 km are selected. The distance between delivery point 1 and delivery point 3 is 60 km, and the average vehicle transportation speed is 30 km / h. Therefore, the transportation tasks of drug 1 and drug 3 are merged. After cost planning, a new route with the lowest cost is obtained, and the transportation tasks are carried out along this new route.

[0111] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0112] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A pharmaceutical supply chain monitoring method based on artificial intelligence, characterized in that: The method comprises the following steps: Step S1. Identify the road transportation network of the pharmaceutical supply area on the GIS map, enter the location of the pharmaceutical supply point and the pharmaceutical inventory, cold chain capacity, and loading and unloading capacity of the supply point, and receive and record pharmaceutical demand information; Step S2. Set the patient's transportation cost weight and transportation distance limit based on the drug demand information, and select a delivery point between the patient and the supply point so that the delivery point is located on the transportation network and the weighted sum of the distance from the supply point to the delivery point and the distance from the patient to the delivery point is minimized; Step S3. For each delivery point, a transport task is created from the supply point to the delivery point. The inter-delivery time between delivery points is calculated through road planning. Transport tasks with inter-delivery times shorter than the difference in the drug's shelf life are merged. Vehicles are arranged according to the transport task, and the transport cost is initialized. Step S4. For each delivery task, a vehicle transport route is planned. The additional cost of each road segment is estimated based on the road network topology, real-time road conditions, traffic restrictions, and travel costs. The additional cost is added to the initial cost to obtain an objective function. This objective function is then planned to minimize transportation costs while meeting the timeliness of the drug, vehicle capacity, and transportation distance constraints. The transport route with the lowest transportation cost is determined. Step S5. Determine the departure time of the vehicle and the patient and the estimated transport duration. If there is excess vehicle capacity, load the most frequently used medications in the area where the last delivery point of the task is located in the excess space and store the medications at the secondary supply point in the area. Step S2 includes: Step S21. Set the patient's transportation cost weight according to the patient's status, satisfying Q1 = k1·Li, where Q1 is the cost of transporting the patient, k1 represents the cost required to transport the patient per unit distance, and Li is the transportation distance from patient i to the delivery point; Step S22. Determine the distance between the supply point and the demand point. Calculate the weighted sum Y of the distance from the supply point to the delivery point and the distance from the patient to the delivery point, satisfying Y = k1·Li + k2·LR, where LR is the distance from the supply point to the delivery point, Li + LR is equal to the distance from the supply point to the patient, and k2 represents the base cost of transporting the drug from the supply point per unit distance. Determine a delivery point on the transportation network that minimizes Y. Step S4 includes: Step S41. Plan all transportation routes from the supply point to the delivery point with a transportation distance below a threshold. Determine the additional cost of each road segment based on the road network topology, real-time road conditions, traffic restrictions, and travel cost planning. The additional cost includes distance cost and travel cost. Add the additional cost to the initial cost to obtain the objective function. Step S42: Plan the minimization objective function: Among them, Q is the objective function, Q0 is the initial transportation cost, k represents the section number of the transportation route, n is the number of sections, c k represents the additional cost of road segment k, t k is the additional time cost of road section k, including the time extension caused by extended distance, speed loss and poor road conditions, w is the delay cost per unit time, Σk1·Li represents the cost of transporting all patients to the delivery points, t0 is the estimated time for the vehicle to arrive at the first delivery point, m is the number of delivery points, tr i is the length of the subsequent transportation to the i-th delivery point, T m is the shelf life of the drug at the mth delivery point, ΣU m is the total volume of drugs, U0 is the vehicle loading capacity, Li is the transportation distance from patient i to the delivery point, L max The upper limit of the distance for transporting patients; Step S43. All transportation routes are planned in succession, and the transportation route that meets the constraints of drug timeliness, vehicle capacity and transportation distance and has the lowest transportation cost is determined for drug transportation.

2. The method for monitoring the pharmaceutical supply chain based on artificial intelligence according to claim 1, characterized in that: Step S1 includes: Step S11. Demarcate the service area of ​​a single supply point based on the Voronoi diagram. Use ArcGIS or QGIS mapping software to enter the regional transportation network within the supply point's service area. Mark the supply point location on the map. Dynamic inbound and outbound warehousing tools record and update the supply point's location, drug inventory, cold chain capacity, and loading and unloading capabilities. Step S12: The cloud platform receives drug supply demands within the service scope and generates drug demand information, which includes patient status, drug demand type, demand quantity, and demand location.

3. The method for monitoring the pharmaceutical supply chain based on artificial intelligence according to claim 2, characterized in that: Step S3 includes: Step S31. Determine the delivery point for each piece of drug demand information within the cycle, and create a transportation task for each delivery point. The transportation task includes: the supply point location, the delivery point location, the type of drug to be transported, and the quantity of the drug; Step S32. Plan the transportation routes between delivery points and determine the continuous transportation time tr between delivery points, satisfying tr=te+Lr / v, where Lr is the length of the transportation route between delivery points, v is the vehicle transportation speed, and te is the time loss, which is set by the congestion level of the transportation route and the time for delivering the medicines.

4. The method for monitoring the pharmaceutical supply chain based on artificial intelligence according to claim 3, characterized in that: Step S5 includes: Step S51. Predict drug usage within the sub-regions of each delivery point. If the vehicle is not fully loaded, additional medication is stored in the remaining space. The medications stored are those with the highest frequency of use in the sub-region of the last delivery point of the mission and whose shelf life is greater than the mission transportation time. Step S52: The extra drugs are stored at the secondary supply point of the sub-region. When the drug demand of the sub-region is less than the demand for the drugs stored at the secondary supply point, the secondary supply point is used to supply and transport the drugs.

5. A pharmaceutical supply chain monitoring system based on artificial intelligence, wherein the system executes the pharmaceutical supply chain monitoring method based on artificial intelligence as claimed in claim 1, characterized in that: The system includes the following modules: a supply scheduling module, a two-way delivery module, a cost management module, a path planning module, and a redundancy utilization module; The supply scheduling module is used to record the location of supply points, drug inventory, cold chain capacity, and loading and unloading capabilities, and to receive real-time drug demand information. The drug demand information includes patient status, drug demand type, demand quantity, and demand location, and is marked on an electronic map to form a drug supply map. For each demand information, the drug shelf life and minimum transportation space are determined based on the drug type; The bidirectional delivery module is used to preset transportation cost weights and transportation distance limits for patients based on demand information, select a delivery point between the patient and the supply point, and the delivery point is located on the transportation network so that the weighted sum of the distance from the transport vehicle to the delivery point and the distance from the patient to the delivery point is minimized. The vehicle waiting time is determined by the difference between the shelf life of the drug and the transportation time from the supply point to the delivery point. The number of dispatches is determined based on the drug transportation space demand, and the transportation tasks with the minimum transportation time between the delivery points being less than the difference in the shelf life of the drugs are merged, and a cost function between the transportation cost and the transportation distance of a single task is output. The cost management module is used to plan the transportation routes from the starting supply point to the delivery point according to the road network topology, real-time road conditions, traffic restrictions and travel costs. Each transportation route corresponds to a transportation plan. The cost function is processed according to the actual distance and additional cost in the transportation plan to obtain the objective function of each transportation plan; The path planning module is used to plan the objective function through a meta-heuristic algorithm to minimize the sum of the drug transportation cost and the patient transportation cost, while meeting the constraints of drug timeliness, vehicle capacity and transportation flow, output the route with the lowest transportation cost, use it as the planned route for vehicle supply, and determine the departure time of the vehicle and the patient and the estimated transportation time; The redundancy utilization module is used to predict the use of medicines in a single area. When the vehicle is not fully loaded, additional medicines are stored in the remaining space. The types of stored medicines are the most frequently used medicines in the area where the last delivery point of the mission is located, and the shelf life is longer than the mission transportation time. The additional medicines are stored at the secondary supply point in the area. When the demand for medicines in the area is less than the demand for storing medicines at the secondary supply point, the secondary supply point will be used for transportation.

6. The artificial intelligence-based pharmaceutical supply chain monitoring system according to claim 5, characterized in that: The supply scheduling module includes: a transportation network unit and a demand input unit; The transportation network unit is used to input a map of the drug supply area through ArcGIS or QGIS map software to identify the road transportation network of the supply area; The demand input unit is used to record the supply point location, drug inventory, cold chain capacity and loading and unloading capacity through dynamic warehousing and storage tools, and receive drug usage demand.

7. The artificial intelligence-based pharmaceutical supply chain monitoring system according to claim 6, characterized in that: The two-way delivery module includes: an intermediate delivery unit, a dispatch planning unit and a cost accounting unit; The intermediate delivery unit is used to divide the service range of the supply point based on the Voronoi diagram and allocate patients to the nearest delivery point; The dispatch planning unit is used to determine the number of vehicles to be dispatched and the dispatch distance, and to combine the transportation routes of different drugs using the Clarke-Wright economizing algorithm; The cost accounting unit is used to initialize the distance cost of drug transportation according to the distance from the supply point to the delivery point to obtain a cost function.

8. The artificial intelligence-based pharmaceutical supply chain monitoring system according to claim 7, characterized in that: The cost management module includes: a transportation route unit and a target cost unit; The transport route unit is used to plan all available transport routes according to the road traffic status and keep the transport distance below a threshold; The target cost unit is used to predict the loss cost and time cost of a vehicle passing through a road section, and accumulate them to obtain an objective function; The path planning module includes: a solution iteration unit and a path selection unit; The solution iteration unit is used to adopt an adaptive ALNS large neighborhood search dynamic adjustment strategy, update the traffic efficiency weight in real time, and plan the objective function; The path selection unit is used to determine a route with the lowest transportation cost and meeting all constraints for transportation.

9. The artificial intelligence-based pharmaceutical supply chain monitoring system according to claim 8, characterized in that: The redundancy utilization module includes: a demand forecasting unit, an additional supply unit and a secondary transportation unit; The demand forecasting unit is used to determine the mean and variance of drug usage frequency according to historical drug supply data and forecast the drug demand range; The additional supply unit is used to introduce a packing problem algorithm to pre-allocate medicines to vehicles so that the space utilization rate of the vehicles reaches a threshold; The secondary transport unit is used to establish a secondary supply system, and when the secondary supply point meets the demand, the secondary supply point will transport the medicine.

Images