Intelligent delivery method for cigarette finished products
By establishing an intelligent delivery method for finished cigarettes and optimizing delivery plans using linear planning and heuristic algorithms, the problems of low efficiency and dynamic interference in the preparation of finished cigarettes are solved, and the optimization of transportation costs and corporate competitiveness are achieved. promote.
Patent Information
- Application Number
- CN202310432895.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-04-21
- Publication Date
- 2024-11-01
AI Technical Summary
The preparation of finished cigarette delivery plans relies on manual labor, which leads to time-consuming, low efficiency, incomplete consideration, and faces uncertainty in sales orders and freight vehicle conditions, resulting in dynamic interference and making it difficult to optimize logistics costs.
Establish an intelligent delivery method for finished cigarettes, and realize automated decision-making by establishing a cost-minimizing delivery model, combining linear planning and heuristic algorithms, combining them with linear planning and heuristic algorithms, taking into account constraints such as transportation time, inventory and vehicle loading capacity.
The cigarette finished product shipment plan has been optimized, the transportation costs have been reduced, the company's core competitiveness has been enhanced, data interconnection and business logic have been achieved, and the delivery plan has been transformed from manual experience to digital control.
Smart Images

Figure BDA0004190927730000021 
Figure BDA0004190927730000023 
Figure BDA0004190927730000026
Abstract
Description
Technical Field
[0001] The present invention relates to the field of logistics, and in particular to an intelligent delivery method for finished cigarette products. Background Art
[0002] With the diversification and personalization of cigarette consumer demand, the demand for finished cigarette logistics presents the characteristics of small batches, multiple varieties and high frequency. The order cycle is shortened, the timeliness is strengthened, and the uncertainty of logistics demand is increased, which puts high demands on the flexible logistics of finished cigarettes. Flexible logistics of finished cigarettes refers to focusing on the logistics needs of customers, responding quickly to customer needs, adjusting logistics operations in a timely manner, and effectively controlling logistics costs, helping enterprises to integrate and optimize logistics resources, and realizing the most optimized delivery mode of finished cigarettes.
[0003] At present, the delivery plan of finished cigarette products is mainly formulated manually, and the delivery plan of finished products is arranged by combining manual experience with multiple constraints. However, in the actual preparation process of the delivery plan of finished cigarette products, there are problems such as long manual scheduling, low efficiency, incomplete consideration of factors, and frequent adjustment of results. At the same time, due to the uncertainty of cigarette sales orders and freight vehicle conditions, the delivery of finished cigarette products is also faced with many dynamic interferences, such as changes in sales orders, freight vehicle maintenance, changes in road conditions, etc., which need to be rescheduled, which has become one of the difficulties in the delivery of finished products of cigarette companies. At present, cigarette logistics companies are in urgent need of an intelligent delivery method to provide a scientific basis for the decision-making of the balance management of multi-warehouse delivery, which can optimize the logistics and transportation costs under the premise of meeting the balance of supply and demand, promote the coordinated operation optimization of finished cigarette storage, delivery and transportation, and market demand push, and promote the gradual transformation of manual experience judgment of delivery plans to digital management and control. Summary of the invention
[0004] In order to further optimize the scheduling of cigarette finished product delivery plans on the premise of meeting market demand, establish a cigarette finished product delivery model, and promote the coordinated operation optimization of finished cigarette warehousing, delivery transportation, and market demand push, so as to achieve the purposes of data interconnection, business logic optimization, more automatic model output, richer transportation types, and more optimized transportation costs, promote the gradual transformation of delivery plans from manual experience judgment to digital management and control, provide data support and scientific basis for decision-making on management work related to supply balance of multi-warehouse delivery, and enhance the core competitiveness of enterprises, the present invention provides an intelligent delivery method for cigarette finished products.
[0005] In order to achieve the above technical purpose, the technical solution of the present invention is:
[0006] A method for intelligently delivering finished cigarette products comprises the following steps:
[0007] Step 1: Taking minimizing the delivery cost as the optimization goal, a delivery model that minimizes the cost includes the full-load delivery cost without carpooling, the carpooling cost, the non-full-load delivery cost without carpooling, and the warehouse transfer cost.
[0008] Step 2: Based on the delivery model, set constraints including delivery transportation time, arrival quantity, delivery quantity, vehicle loading capacity, transfer quantity, arrival time, transfer transportation time and inventory;
[0009] Step three, solve the model: first, consider only the full-load shipping cost and warehouse transfer cost without carpooling, and get the preliminary shipping result based on linear programming; then, based on the heuristic algorithm, add the carpooling cost and the non-carpooling part-full-load shipping cost, optimize the preliminary shipping result, and get the final shipping result.
[0010] The method described, in step 1, the full-load delivery cost D N for:
[0011]
[0012] Where T represents the set of natural days in a month, t d represents a natural day in T, t d ∈T={1,…,|T|};W represents the set of shipping points, w represents a shipping point in W, w∈W={1,…,|W|};L represents the set of arrival points, l,l* represents an arrival point in L, l,l * ∈L={1,...,|L|};I represents the set of vehicle models, i represents a vehicle model in I, i∈I={1,...,|I|};p wl It represents the unit transportation cost from the shipping point w to the arrival point l; Indicates the tth d The number of vehicle types i that are sent from the shipping point w to the arrival point l without carpooling when the vehicle is fully loaded on the day; L i Indicates the loading capacity corresponding to vehicle type i;
[0013] Carpooling cost D S for:
[0014]
[0015] in Indicates that the arrival point l arrives at another arrival point l * The unit transportation cost, Indicates that the arrival point l arrives at another arrival point l * Transportation distance;
[0016] Cost of non-carpooling and non-full-load delivery D M for:
[0017]
[0018] in Indicates the tth d The number of vehicle types i shipped from the shipping point w to the arrival point l without carpooling (not fully loaded, at most one vehicle) every day, and
[0019] Transfer cost D Q for:
[0020]
[0021] in Indicates the tth d Day delivery point w is moved to another delivery point w * The number of vehicle models i, w * ∈W={1,...,|W|}(Since the quantity of warehouse transfer is not as strict as the delivery order, it is better to strive for full load, without carpooling);
[0022] The delivery model is thus established as:
[0023] Min(D N )
[0024] Min(D S )
[0025] Min(D M )
[0026] Min(D Q ).
[0027] In the method, in step 2, the constraints include:
[0028] 1) Shipping time constraints
[0029]
[0030] where dt wl It represents the transportation time from the shipping point to the delivery point; v represents the average driving speed; buffer represents the shipping buffer time;
[0031] 2) Each brand’s t d The arrival demand per day is equal to (t d -dt wl ) before the time
[0032]
[0033] 3) No. t dThe loading capacity of the vehicle that does not share a vehicle from delivery point w to the destination l is greater than or equal to the total shipment volume of delivery point w.
[0034]
[0035] 4) No. t d Day delivery point w is moved to another delivery point w * The vehicle loading capacity used is equal to the total number of products transferred
[0036]
[0037] 5) Control the arrival time to avoid arriving too far in advance. Do not arrive more than 2 days in advance.
[0038]
[0039] t l Indicates the arrival time of a certain brand required by the delivery point l under a certain order.
[0040] 6) Time constraints for warehouse transfer and transportation
[0041]
[0042] 7) Inventory Constraints
[0043]
[0044] Inventory Constraint Meaning
[0045] 8) The total inventory of each delivery point at any time cannot exceed the maximum inventory
[0046]
[0047] 9) The brand inventory value of each delivery point at any time must be greater than or equal to the preset safety stock
[0048]
[0049] The method is characterized in that, in step 3, the preliminary delivery result is obtained through the following process:
[0050] When only the full-load delivery cost and warehouse transfer cost without carpooling are considered, the objective function is:
[0051] min(D N +D Q )
[0052] Then substitute the model and solve it through linear programming to obtain The value of .
[0053] In the method, in step 3, the process of optimizing the preliminary delivery result includes:
[0054] First, optimize the vehicle's underload: According to the results of linear programming Allocate each shipping brand to each vehicle, where the arrangement of the shipping brands is from the brand with the largest demand to the brand with the smallest demand, so that the brand with the largest demand is arranged first; when allocating to each vehicle, arrange the model with the largest loading capacity first, and then arrange the model with the smallest loading capacity, so that the model that is not fully loaded in the end is the model with the smallest loading capacity;
[0055] Then, based on the results of the optimization of vehicles that are not fully loaded, the vehicles that are not fully loaded are included in the carpooling plan, and the carpooling plan is arranged to obtain the final delivery result.
[0056] The method described herein, wherein the carpooling plan arrangement comprises:
[0057] First, consider carpooling with multiple arrival points from the same delivery point:
[0058] ① Classify the delivery plans with the same operator as the destination, so as to establish a pool of cars that can be shared;
[0059] ② For a carpooling set, starting from one of the delivery points, each time search for the nearest arrival point within the predetermined range, with a difference of more than 24 hours in arrival time, a total delivery volume not exceeding the vehicle's full load, and a distance of less than 500 kilometers as the next destination, until the total delivery volume equals the vehicle's load capacity. Each loading will default to using the currently idle vehicle with the largest load capacity, and record the carpooling route and the quantity of each brand sent to each arrival point;
[0060] ③. Repeat step ② until there are no more arrival points for carpooling in the carpooling set;
[0061] Secondly, consider the case where the same arrival point corresponds to multiple delivery points:
[0062] a) Classify the delivery plans with the same delivery point as the operator and build a carpooling set;
[0063] b) For the set of carpoolable vehicles, start from one of the arrival points in the opposite direction, and search for the arrival point within the predetermined range that is closest, has a difference of more than 24 hours in arrival time, and the total shipment volume does not exceed the vehicle's full load as the previous destination each time, until the total shipment volume is greater than the vehicle's loading capacity. Each loading will default to using the vehicle with the largest loading capacity that is currently available, and record the carpooling route and the number of each brand loaded at each delivery point.
[0064] c) Repeat step b) until there are no more arrival points that can be shared in the shared vehicle set.
[0065] Finally, the remaining delivery plans will be considered as non-carpooling and non-full-load delivery.
[0066] The technical effect of the present invention is that the present invention combines the needs and characteristics of finished cigarette logistics, and studies the intelligent delivery of finished cigarette logistics based on the data collected and integrated in real time by the enterprise information system. On the premise of meeting market demand, the delivery plan arrangement of finished cigarettes is further optimized, a finished cigarette delivery model is established, and the coordinated operation optimization of finished cigarette storage, delivery transportation, and market demand push is promoted, so as to achieve the purpose of data interconnection, business logic optimization, more automatic model output, richer transportation types, and more optimized transportation costs, promote the delivery plan from manual experience judgment to digital management and control, provide data support and scientific basis for the decision-making of management work related to the balance of supply of multi-warehouse delivery, and enhance the core competitiveness of enterprises. DETAILED DESCRIPTION
[0067] Data source acquisition
[0068] The data sources of various indicators involved in the present invention are all from various production systems or manual ledgers of the enterprise. Relevant delivery data are extracted from the CRM system, ERP system, and logistics system. The data is integrated into the enterprise big data platform through the ETL method. A basic data maintenance table for the finished product logistics delivery model is established in the big data platform, and data processing and analysis are unified. The main data requirements are shown in the following table.
[0069] Table 1 Main data requirements for finished product logistics delivery
[0070]
[0071]
[0072] 2. Model for delivery of finished cigarettes
[0073] The model is established as follows:
[0074] (1) Parameter definition
[0075] gather:
[0076] W: the set of shipping points, w, w * Both represent the shipping point, and w, w * ∈W={1,...,|W|}.
[0077] L: the set of arrival points, l, l * ∈L={1,...,|L|}.
[0078] P: the set of brands, p∈P={1,...,|P|}.
[0079] T: the set of natural days in a month, t d ∈T={1,…,|T|},t d-1 That is t d The day before.
[0080] I: The set of vehicle models, i∈I={1,...,|I|}.
[0081] B: A set of orders, b∈B={1,...,|B|}.
[0082] Vehicle Information:
[0083] v: average driving speed, which defaults to 50 kilometers per hour in this embodiment.
[0084] L i : Loading capacity corresponding to vehicle type i.
[0085] Shipping distance and cost:
[0086] p wl : The unit transportation cost from the shipping point w to the arrival point l.
[0087] Shipping point w is moved to another shipping point w * unit transportation cost.
[0088] From one arrival point to another arrival point * Unit transportation cost (used in carpooling scenarios).
[0089] d wl : The transportation distance from the shipping point w to the arrival point l.
[0090] From delivery point w to another delivery point w * The transfer transportation distance.
[0091] From one arrival point to another arrival point * Transportation distance (used in carpooling scenarios).
[0092] Stock Information:
[0093] SI wp : Real-time inventory of brand p in shipping point w.
[0094] C w : The maximum storage capacity of the shipping point w.
[0095] No. d Days to ship from point w of brand p's inventory.
[0096] Additional Information:
[0097] No. d The quantity of brand p produced at shipping point w per day.
[0098] No. d The arrival demand of brand p at arrival point l on day.
[0099] No. d The number of repurchases for brand p at delivery point w per day.
[0100] buffer: Shipping buffer time, mainly comes from driver rest, loading, unloading, weather reasons, etc.
[0101] SS p :Safety stock considerations for brand p at each delivery point.
[0102] (2) Model decision variables
[0103] No. d The quantity of brand p shipped from shipping point w to arrival point l per day.
[0104] b Order No. t d The quantity of brand p shipped from shipping point w to arrival point l per day.
[0105] No. d Day delivery point w is moved to another delivery point w * The number of brands p.
[0106] No. d Day delivery point w is moved to another delivery point w * The number of vehicle models i, because the quantity of warehouse transfer is not as strict as the delivery order, it is better to strive for full load, without carpooling. That is, the calculation is based on the full load of the vehicle during the warehouse transfer process, and the situation of not being fully loaded is not considered.
[0107] No. d The number of vehicle type i sent from shipping point w to arrival point l without carpooling (fully loaded) on that day.
[0108] No. dThe number of vehicle types i that are sent from the shipping point w to the arrival point l without carpooling (not fully loaded, at most one vehicle). The reason why only one vehicle is not fully loaded is that for the same operator, there must be only one vehicle that is not fully loaded, because when loading, the goods will be filled in sequence according to the order of the vehicles. If there is a situation of not being fully loaded, only the last vehicle will not be fully loaded. In this embodiment, the operators are generally divided by province. For example, there must be only one vehicle that is not fully loaded when sent to Jilin Province, and there may also be one vehicle that is not fully loaded when sent to Qinghai Province.
[0109] Brand p in t d The full-load shipment volume of vehicle type i from shipping point w to arrival point l on that day
[0110] Brand p in t d The number of incomplete shipments from the shipping point w to the arrival point l on that day No. d The number of vehicle types i that are sent from the shipping point w to the arrival point l by carpooling on the day
[0111] No. d Day delivery point w to delivery point l whether to share a car
[0112] No. d The number of vehicle types i that are sent from the delivery point w to the arrival point l without carpooling
[0113] 1) Full-load delivery cost without carpooling
[0114]
[0115] 2) Carpooling costs
[0116]
[0117] 3) Cost of non-carpooling and non-full-load delivery
[0118]
[0119] 4) Transfer cost:
[0120]
[0121] (3) Objective function and constraints
[0122] Based on minimizing cost as the optimization goal, the mathematical model for cigarette finished product delivery is established as follows:
[0123] Min(D N )
[0124] Min(D S )
[0125] Min(D M )
[0126] Min(D Q )
[0127] Constraints:
[0128] 1) Shipping time constraints
[0129]
[0130] where dt wl Indicates the transportation time from the shipping point to the arrival point.
[0131] 2) Each brand’s t d The arrival demand per day is equal to (t d -dt wl ) before the time
[0132]
[0133] 3) No. t d The loading capacity of the vehicle that does not share a vehicle from delivery point w to the destination l is greater than or equal to the total shipment volume of delivery point w.
[0134]
[0135] 4) No. t d Day delivery point w is moved to another delivery point w * The vehicle loading capacity used is equal to the total number of products transferred
[0136]
[0137] 5) Control the arrival time to avoid arriving too far in advance. Do not arrive more than 2 days in advance.
[0138]
[0139] t l Indicates the arrival time of a certain brand required by the delivery point l under a certain order.
[0140] 6) Time constraints for warehouse transfer and transportation
[0141]
[0142] 7) Inventory Constraints
[0143]
[0144] Inventory Constraint Meaning
[0145] 8) The total inventory of each delivery point at any time cannot exceed the maximum inventory
[0146]
[0147] 9) The brand inventory value of each delivery point at any time must be greater than or equal to the preset safety stock
[0148]
[0149] 3. Model solution
[0150] This model is solved by designing a linear optimization algorithm combined with a heuristic algorithm. The algorithm is divided into two stages. In the first stage, the linear optimization algorithm is used to search and obtain preliminary results. In the second stage, the heuristic algorithm is used to optimize the preliminary results and obtain the final production scheduling results.
[0151] (1) Linear programming solution
[0152] In the first stage, only the two cases of non-carpooling and warehouse transfer are considered, and the objective function is:
[0153] min(D N +D Q )
[0154] Substitute the model and solve it through linear programming to get The present embodiment directly calls the linear programming package pulp in Python to solve the problem. In the specific implementation, other common linear programming algorithms may be considered to find the extreme value to obtain the corresponding solution.
[0155] (2) Solve the model by writing a heuristic algorithm
[0156] Based on the initial production scheduling results obtained by solving the linear programming algorithm, constraints such as brand upper and lower limit repair, batch constraint repair, peak season production mode repair, brand production sequence, etc. are incorporated into the heuristic algorithm for solving to optimize the production scheduling results. The algorithm logic details are as follows.
[0157] ① Repair if the vehicle is not fully loaded:
[0158] According to the results of linear programming Assign each shipping brand to each vehicle.
[0159] 1. (t d The number of brand p sent from shipping point w to arrival point l on that day is sorted from large to small, with brands with greater demand given priority.
[0160] 2. Give priority to vehicles with large loading capacity to ensure that the remaining vehicles that are not fully loaded are small vehicles to reduce subsidy costs.
[0161] ②Carpool repair:
[0162] 1. Select the vehicles that are not fully loaded at the end and add them to the carpooling consideration.
[0163] 2. First consider carpooling for multiple arrival points from the same delivery point
[0164] a) Classify the delivery plans with the same operator as the destination into one category, that is, they can be considered for carpooling, and construct a carpooling set. That is, all the partially loaded vehicles with the same operator as the destination are classified into this carpooling set.
[0165] b) For a carpooling collection, starting from the same delivery point, each time search for the nearest arrival point within 500 kilometers, with a difference of more than 24 hours in arrival time and a total delivery volume not exceeding the vehicle's full load as the next destination, until the total delivery volume exceeds the vehicle's load capacity (the default is to use the vehicle with the largest load capacity at the beginning, and the vehicle model will be adjusted based on the actual transportation volume later), and record the carpooling route and the quantity of each brand sent to each arrival point.
[0166] c) Repeat step b until there are no more arrival points that can be shared in the shared car set.
[0167] 3. Secondly, consider the carpooling situation where the same arrival point corresponds to multiple delivery points
[0168] a) Classify the delivery plans of the same operator as one category, that is, they can be considered for carpooling, and construct a carpooling set. That is, all the partially loaded vehicles with the same operator as the delivery point are classified into this carpooling set.
[0169] b) For a group of carpooling vehicles, start from the same arrival point in the opposite direction, and search for the nearest arrival point within 500 kilometers each time, with a difference of more than 24 hours in arrival time and a total shipment volume not exceeding the vehicle's full load as the previous destination, until the total shipment volume exceeds the vehicle's loading capacity (the vehicle with the largest loading capacity is used by default at the beginning, and the vehicle model is adjusted according to the actual transportation volume). The carpooling route and the number of each brand loaded at each delivery point are recorded.
[0170] c) Repeat step b until there are no more arrival points that can be shared in the shared car set.
[0171] Finally, the remaining delivery plans will be considered as non-carpooling and non-full-load delivery.
Claims
1. An intelligent method for delivering finished cigarette products, characterized in that: The following steps are involved: Step 1: Taking minimizing the delivery cost as the optimization goal, a delivery model that minimizes the cost includes the full-load delivery cost without carpooling, the carpooling cost, the non-full-load delivery cost without carpooling, and the warehouse transfer cost. Step 2: Based on the delivery model, set constraints including delivery transportation time, arrival quantity, delivery quantity, vehicle loading capacity, transfer quantity, arrival time, transfer transportation time and inventory; Step three, solve the model: first, consider only the full-load shipping cost and warehouse transfer cost without carpooling, and get the preliminary shipping result based on linear programming; then, based on the heuristic algorithm, add the carpooling cost and the non-carpooling part-full-load shipping cost, optimize the preliminary shipping result, and get the final shipping result.
2. The method according to claim 1, characterized in that In step 1, the cost of full-load delivery without carpooling is D N for: Where T represents the set of natural days in a month, t d represents a natural day in T, t d ∈T={1,…,|T|};W represents the set of shipping points, w represents a shipping point in W, w∈W={1,…,|W|};L represents the set of arrival points, l,l * represents an arrival point in L, l,l * ∈L={1,...,|L|};I represents the set of vehicle models, i represents a vehicle model in I, i∈I={1,...,|I|};p wl It represents the unit transportation cost from the shipping point w to the arrival point l; Indicates the tth d The number of vehicle types i that are sent from the shipping point w to the arrival point l without carpooling when the vehicle is fully loaded on the day; L i Indicates the loading capacity corresponding to vehicle type i; Carpooling cost D S for: in Indicates that the arrival point l arrives at another arrival point l * The unit transportation cost, Indicates that the arrival point l arrives at another arrival point l * Transportation distance; Cost of non-carpooling and non-full-load delivery D M for: in Indicates the tth d The number of vehicle type i that is sent from the delivery point w to the arrival point l without carpooling, and Transfer cost D Q for: in Indicates the tth d Day delivery point w is moved to another delivery point w * The number of vehicle models i, w * ∈W={1,...,|W|}; The delivery model is thus established as: Min(D N ) Min(D S ) Min(D M ) Min(D Q )。 3. The method according to claim 2, characterized in that In the step 2, the constraints include: 1) Shipping time constraints where dt wl It represents the transportation time from the shipping point to the delivery point; v represents the average driving speed; buffer represents the shipping buffer time; 2) Each brand’s t d The arrival demand per day is equal to (t d -dt wl ) before the time 3) No. t d The loading capacity of the vehicle that does not share a vehicle from delivery point w to the destination l is greater than or equal to the total shipment volume of delivery point w. 4) No. t d Day delivery point w is moved to another delivery point w * The vehicle loading capacity used is equal to the total number of products transferred 5) Control the arrival time to avoid arriving too far in advance. Do not arrive more than 2 days in advance. t l Indicates the arrival time of a certain brand required by the delivery point l under a certain order. 6) Time constraints for warehouse transfer and transportation 7) Inventory Constraints Inventory Constraint Meaning The amount of warehouse transfers before time t d Shipment quantity per day - tth d (the amount of inventory shipped out of the warehouse per day). 8) The total inventory of each delivery point at any time cannot exceed the maximum inventory 9) The brand inventory value of each delivery point at any time must be greater than or equal to the preset safety stock 4. The method according to claim 3, characterized in that In step 3, the preliminary delivery result is obtained through the following process: When only the full-load delivery cost and warehouse transfer cost without carpooling are considered, the objective function is: min(D N +D Q ) Then substitute the model and solve it through linear programming to obtain The value of .
5. The method according to claim 4, characterized in that In step 3, the process of optimizing the preliminary delivery results includes: First, optimize the vehicle's underload: According to the results of linear programming Allocate each shipping brand to each vehicle, where the arrangement of the shipping brands is from the brand with the largest demand to the brand with the smallest demand, so that the brand with the largest demand is arranged first; when allocating to each vehicle, arrange the model with the largest loading capacity first, and then arrange the model with the smallest loading capacity, so that the model that is not fully loaded in the end is the model with the smallest loading capacity; Then, based on the results of the optimization of vehicles that are not fully loaded, the vehicles that are not fully loaded are included in the carpooling plan, and the carpooling plan is arranged to obtain the final delivery result.
6. The method according to claim 5, characterized in that The carpooling program includes: First, consider carpooling with multiple arrival points from the same delivery point: ① Classify the delivery plans with the same operator as the destination, so as to establish a pool of cars that can be shared; ② For a carpooling set, starting from one of the delivery points, each time search for the nearest arrival point within the predetermined range, with a difference of more than 24 hours in arrival time, a total delivery volume not exceeding the vehicle's full load, and a distance of less than 500 kilometers as the next destination, until the total delivery volume equals the vehicle's load capacity. Each loading will default to using the currently idle vehicle with the largest load capacity, and record the carpooling route and the quantity of each brand sent to each arrival point; ③. Repeat step ② until there are no more arrival points for carpooling in the carpooling set; Secondly, consider the case where the same arrival point corresponds to multiple delivery points: a) Classify the delivery plans with the same delivery point as the operator and build a carpooling set; b) For the set of carpoolable vehicles, start from one of the arrival points in the opposite direction, and search for the arrival point within the predetermined range that is closest, has a difference of more than 24 hours in arrival time, and the total shipment volume does not exceed the vehicle's full load as the previous destination each time, until the total shipment volume is greater than the vehicle's loading capacity. Each loading will default to using the vehicle with the largest loading capacity that is currently available, and record the carpooling route and the number of each brand loaded at each delivery point. c) Repeat step b) until there are no more arrival points that can be shared in the shared vehicle set. Finally, the remaining delivery plans will be considered as non-carpooling and non-full-load delivery.
Citation Information
Cited By
Finished cigarette warehouse net layout optimization method and system based on multi-model collaboration
CN120410409A
