A logistics delivery vehicle scheduling and planning method and system

Abstract

The present invention proposes a scheduling planning method and system for logistics pickup vehicles, which relates to the technical field of logistics pickup vehicle scheduling. First, the logistics virtual sub-areas are divided into sub-areas, and each logistics virtual sub-area is equipped with its own pickup vehicles, which fundamentally guarantees the efficiency of picking up goods within the sub-areas of the logistics virtual sub-areas, increases the number of pickups by own pickup vehicles, and controls costs. Then, considering the instability of the actual shipping period of the consignor, the pickup time is predicted based on the consignor's historical order pickup data, which reduces manual communication costs and reduces logistics pickup costs and improves the timeliness of pickup by improving the timeliness of pickup.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics pickup vehicle scheduling, and more specifically, to a logistics pickup vehicle scheduling planning method and system. Background Art

[0002] In the field of large-ticket logistics, when picking up goods from shippers, the dispatcher communicates with the shipper about the pick-up time period, and allocates vehicles to pick up goods based on multiple factors such as the loading of the owned vehicles, the allocable time period, the logistics center where the vehicles are diverted, etc., or rents external vehicles to pick up goods.

[0003] At present, in order to improve the intelligence of vehicle mobilization during logistics pick-up and delivery, thereby improving the efficiency of logistics pick-up and delivery and reducing the cost of cargo transportation, many technical personnel in this field have made efforts in this regard. For example, the prior art discloses a method and system for optimal multi-vehicle distribution, which takes into account the number of transportations, past order volume, and order congestion level. If the preset area has a total order volume of not less than a predetermined value each month, the preset area is set as an overlapping transportation area, allowing joint transportation from other adjacent areas, and optimizing the distribution of multiple vehicles so that suitable vehicles can carry out cargo transportation within the predetermined transportation area to minimize the total transportation cost of each transportation area.

[0004] However, in the field of large-ticket logistics, the actual delivery time of small and medium-sized enterprise shippers is uncertain due to the uncertainty of actual production time. Therefore, the time and weight of the shippers' release of the source of goods are random. At the same time, the geographical distribution of the source of goods is relatively scattered, and dispatchers generally need to participate manually. In addition, there are many variable factors to be considered. It is difficult to effectively dispatch vehicles to pick up goods based on the above ideal optimization method, and it is difficult to control the cost and timeliness of picking up goods. Summary of the Invention

[0005] In order to solve the problem that the existing method of dispatching vehicles to pick up goods is difficult to adapt to the multiple variable factors of large-scale logistics, the present invention proposes a scheduling planning method and system for logistics pickup vehicles, which rationally dispatches a limited number of self-owned pickup vehicles and combines them with renting external vehicles to reduce logistics pickup costs and improve pickup timeliness.

[0006] In order to achieve the above technical effects, the technical solutions of the present invention are as follows:

[0007] A method for scheduling and planning logistics delivery vehicles, the method comprising:

[0008] S1. Divide logistics into virtual sub-divisions and equip each virtual sub-division with its own delivery vehicles;

[0009] S2. Estimate the daily order volume for each logistics virtual division area, and allocate the company's own pick-up vehicles according to the estimated daily order volume of each logistics virtual division area.

[0010] S3. Predict the actual pick-up time and the list of available pick-up time periods based on the historical order pick-up data of the shipper.

[0011] S4. Shipper places an order: Based on the number of days the shipper places an order within a week, combined with the actual pick-up time and the list of available pick-up time periods predicted in S3, predict the actual pick-up time and the list of available pick-up time periods for the current shipper, and determine the logistics virtual division area to which the shipper belongs in combination with S1.

[0012] S5. Determine whether the company's own pick-up vehicles equipped within the logistics virtual division area to which the shipper belongs can pick up the goods within the actual pick-up time and the list of available pick-up time periods of the current shipper. If so, dispatch the company's own pick-up vehicles to pick up the goods; otherwise, search for neighboring vehicles within the configuration range for pick-up dispatching, and execute step S6.

[0013] S6. If the neighboring vehicles cannot be used for pick-up dispatching, then match and evaluate the company's own pick-up vehicles in other logistics virtual division areas. If the matching and evaluation are passed, arrange for the company's own pick-up vehicles to perform pick-up dispatching; otherwise, compare the cost of renting external vehicles with the cost of arranging the company's own pick-up vehicles in other logistics virtual division areas, and select the one with the lower cost as the vehicle for picking up the goods and perform the pick-up.

[0014] In this technical solution, first, divide the logistics virtual division areas. Equip each logistics virtual division area with the company's own pick-up vehicles, which basically guarantees the pick-up efficiency within the scope of the logistics virtual division areas, increases the pick-up quantity of the company's own pick-up vehicles, and controls costs; then, considering the unstable actual shipping time periods of the shippers, predict the pick-up time based on the historical order pick-up data of the shippers, reduce the manual communication cost, and through reasonable dispatching of the limited number of the company's own pick-up vehicles and in combination with renting external vehicles, reduce the logistics pick-up cost and improve the pick-up timeliness.

[0015] Preferably, in step S1, divide the logistics virtual division areas based on the geographical location of the shippers' shipments. When dividing, obtain the historical order pick-up data of the shippers, extract features from the historical order pick-up data of the shippers. The features include: the number of active shippers, the number of orders, the weight of the goods, and the order placement cycle. Then, perform clustering operations in combination with machine learning algorithms, and divide the geographical location of each shipper's shipment into c1*w1 logistics virtual division areas, where c1 is the number of the company's own pick-up vehicles configured for a certain logistics virtual division area, w1 is a configuration parameter, and 0 < w1 <= 1; after completing the division of the logistics virtual division areas, trim the logistics virtual division areas for efficient pick-up dispatching.

[0016] Preferably, in step S2, the estimate is made by taking into account the week-on-week changes, holidays, and promotional periods. The daily order quantity estimate includes the non-holiday order quantity estimate and the holiday order quantity estimate. The estimate formula for the non-holiday order quantity y1 is:

[0017] y1=(c2*w2+c3*(1-w2))*c4 / c5*w3

[0018] Among them, c2 represents the daily order volume for the same day of the same week last month; c3 represents the daily order volume for the same day of the previous week; the month-on-month change in the daily order volume estimate includes both monthly and weekly changes; w2 represents the proportion of the monthly change in the daily order volume estimate, which is a configured value, 0 <= w2 <= 1; c4 represents the total order volume for the past 7 days; c5 represents the total order volume for the corresponding 7 days of c4 in the same period last month; and w3 represents the influencing factor of the promotion cycle.

[0019] The estimated formula for holiday order quantity y2 is:

[0020] y2=c6 / c7*c8*(1+w4)

[0021] Among them, c6 represents the total daily order volume in the 7 days before the statutory holiday in this cycle; c7 represents the total daily order volume in the 7 days before the same statutory holiday in the previous cycle; c8 represents the daily order volume on the same statutory holiday in the previous cycle; w4 represents the holiday price adjustment coefficient, -1 <w4<1;

[0022] The formula for allocating self-owned pickup vehicles based on the estimated daily order volume of each virtual logistics branch area is:

[0023]

[0024] Among them, x p represents the self-owned pickup vehicle allocated to the pth logistics virtual branch area, y i p It represents the estimated daily order volume of the p-th logistics virtual branch area. On non-holidays, i is 1, and on holidays, i is 2. w6 represents the total number of self-owned pickup vehicles configured.

[0025] Preferably, in step S3, characteristic elements are obtained from the consignor's historical order pickup data, and the characteristic elements include: the consignor's order time, the expected pickup time, and the actual pickup time. The process of predicting the actual available pickup time and the available pickup time period list based on the consignor's historical order pickup data includes:

[0026] S31. Calculate the mean square error between the actual pickup time and the expected pickup time and perform normalization.

[0027] Among them, the calculation formula of mean square error f is:

[0028]

[0029] in, The mean of the difference between the actual delivery time and the expected delivery time of the consignor, d p It represents the difference between the actual pickup time and the expected pickup time for a single order on a specific day of the week. The normalized formula is:

[0030]

[0031] Among them, f min represents the minimum value of the mean square error, f mix represents the maximum value of the mean square error, and g represents the value after normalization;

[0032] S32. Predict the delay time t of the actual delivery time of the consignor, expressed as:

[0033]

[0034] Among them, r is the actual delivery time of the order corresponding to the historical Sunday of the consignor's historical order, h is the expected delivery time of the order corresponding to the historical Sunday of the consignor's historical order, n is the total number of orders, j is a configuration parameter, and the default value is 2;

[0035] S33. Add the expected delivery time of the order on that day to the delay time t of the actual delivery time of the consignor to obtain the predicted actual delivery time;

[0036] S34. A list of available pickup time periods is obtained through statistics of the actual pickup time of the shipper's historical orders. If the shipper has a history of picking up goods during the corresponding time period, the corresponding time period is added to the list of available pickup time periods. This is used to select the next available pickup time period when the goods cannot be picked up during the predicted time period.

[0037] Preferably, in step S4, when judging whether the owned pickup vehicle equipped in the logistics virtual branch area to which the consignor belongs can pick up the goods within the current consignor's actual pickup time and pickup time period list, the owned pickup vehicle allocated to the logistics virtual branch area is determined according to the logistics virtual branch area to which the consignor belongs, and a search is performed to find out whether there is a vehicle that can pick up the goods in the corresponding actual pickup time and pickup time period list. If there is a owned pickup vehicle that can pick up the goods within the current consignor's actual pickup time and pickup time period list, the owned pickup vehicle is dispatched to pick up the goods, and a vehicle pickup record is created, including order information, pickup time period and vehicle.

[0038] Preferably, if there is no self-owned pickup vehicle that can pick up the goods within the current consignee's actual pickup time and pickup period list, the adjacent vehicle is searched within the configuration range. The adjacent vehicle refers to a vehicle that is idle during the predicted pickup period, or is within the pickup period and pickup area configuration range. The configuration range is the first configuration range. The first configuration range takes the center of the logistics virtual branch area to which the consignee belongs as the center of the circle, and the radius is 5km.

[0039] Preferably, in step S6, the process of matching and evaluating the self-owned delivery vehicles in other logistics virtual division areas is as follows:

[0040] S61. Check if the owned pickup vehicles in other virtual logistics divisions are available within the current shipper's actual available pickup time and available pickup time slots, and are within a second configuration range. The second configuration range is based on the center of the previous order pickup location and the next order pickup location as reference points, and has a radius greater than the radius of the first configuration range.

[0041] S62. If the average number of pickup orders per vehicle per hour parameter w7 is configured within the second configuration range, use the following formula to evaluate whether the self-owned pickup vehicles in other logistics virtual division areas can be used for pickup. The formula is:

[0042] rd=(r1-r2) / r3 / (r4-r5)

[0043] Where rd represents the estimated average number of pickup orders per vehicle per hour, r1 represents the estimated number of orders for the day corresponding to other virtual logistics divisions, r2 represents the actual number of orders for the day corresponding to other virtual logistics divisions, r3 represents the number of self-owned pickup vehicles bound to other virtual logistics divisions, r4 represents the order cut-off hours for the day, and r5 represents the actual available pickup hours at that time.

[0044] If rd>=w7, then the self-owned pickup vehicles in other virtual logistics division areas cannot be used to pick up goods. Otherwise, the self-owned pickup vehicles in other virtual logistics division areas can be used to pick up goods.

[0045] S63. If outside the second configuration range, compare the cost of renting an external vehicle with the cost of arranging a self-owned pickup vehicle for another virtual logistics branch area. The cost calculation expression for arranging a self-owned pickup vehicle for another virtual logistics branch area is:

[0046] cost1=k1+k2*2*k3 / 100*L

[0047] Where cost1 represents the cost of arranging self-owned pickup vehicles in other virtual logistics divisions; k1 represents the average fixed cost of a single pickup by a self-owned pickup vehicle; k2 represents the distance traveled for the pickup; k3 represents the average fuel consumption per 100 kilometers of the self-owned pickup vehicle; and L represents the fuel price per liter.

[0048] The cost calculation expression for renting an external vehicle is:

[0049] cost2=k4+(k2*2-k5)*q

[0050] Where cost2 represents the cost of renting an external vehicle; k2 represents the distance traveled to pick up the goods; k4 represents the starting price for renting different grades of vehicles according to the weight of the goods; k5 represents the number of kilometers rented by the external vehicle, and q represents the unit price for renting an external vehicle for additional kilometers.

[0051] When cost1>=cost2, arrange for self-owned pickup vehicles in other logistics virtual branch areas to pick up the goods; otherwise, rent external vehicles to pick up the goods.

[0052] Preferably, after step S6, a pick-up plan for logistics pick-up vehicles is generated, and the pick-up plan includes a pick-up time period and a vehicle for picking up the goods. The dispatcher notifies the consignee via text message and notifies the consignee to reserve a pick-up time period and a vehicle for picking up the goods according to the pick-up plan; when the consignee changes the pick-up time period, the dispatcher changes the pick-up time period to an idle time period that can be reserved for the vehicle, and the idle time period that can be reserved for the owned pick-up vehicle is extracted from the scheduling plan of the owned pick-up vehicle. The idle time period that can be reserved for the external vehicle is reassigned once, and the stage of rescheduling the pick-up is when the consignee finally confirms the pick-up time period.

[0053] Preferably, after picking up the goods, the cargo loading is completed and the vehicles are diverted. After the diversion is completed, the owned pickup vehicles return to their respective virtual logistics branch areas and wait for the next pickup scheduling.

[0054] The present invention also proposes a logistics delivery vehicle scheduling and planning system, comprising:

[0055] Virtual division configuration module, used to divide logistics virtual division areas and equip each logistics virtual division area with its own delivery vehicles;

[0056] The order volume estimation configuration module is used to estimate the daily order volume of each virtual logistics branch area and allocate self-owned pickup vehicles based on the estimated daily order volume of each virtual logistics branch area;

[0057] The delivery forecast module predicts the actual delivery time and delivery period based on the cargo owner's historical order delivery data;

[0058] The consignor module is used for consignor orders: Based on the number of days the consignor places an order within a week, combined with the actual available pickup time and available pickup time period predicted by the pickup prediction module, the current consignor's actual available pickup time and available pickup time period are predicted, and combined with the virtual division configuration module, the consignor's logistics virtual division area is determined;

[0059] The judgment module is used to determine whether the self-owned pickup vehicles deployed in the virtual logistics branch area to which the cargo owner belongs can pick up the cargo within the current cargo owner's actual available pickup time and available pickup time list. If so, the self-owned pickup vehicle is dispatched to pick up the cargo; otherwise, a nearby vehicle is searched within the configured range for pickup scheduling;

[0060] The matching and evaluation module matches and evaluates the self-owned pickup vehicles in other virtual logistics division areas when neighboring vehicles cannot be scheduled for pickup. If the matching evaluation passes, the self-owned pickup vehicle is arranged for pickup. Otherwise, the cost of renting an external vehicle is compared with the cost of arranging the self-owned pickup vehicles in other virtual logistics division areas, and the vehicle with the lower cost is selected as the vehicle for pickup.

[0061] Compared with the prior art, the beneficial effects of the technical solution of the present invention are:

[0062] The present invention proposes a scheduling planning method and system for logistics pick-up vehicles. First, the logistics virtual sub-areas are divided into sub-areas, and each logistics virtual sub-area is equipped with its own pick-up vehicles, which fundamentally guarantees the efficiency of picking up goods within the sub-areas of the logistics virtual sub-areas, increases the number of pick-up goods by own pick-up vehicles, and controls the cost. Then, considering the instability of the actual shipping period of the consignor, the pick-up time is predicted based on the consignor's historical order pick-up data, which reduces the cost of manual communication and reduces the logistics pick-up cost by reasonably scheduling a limited number of own pick-up vehicles and combining them with leasing external vehicles. The timeliness of pick-up is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic diagram showing a flow chart of the method for scheduling and planning logistics delivery vehicles proposed in Example 1 of the present invention;

[0064] Figure 2 A schematic diagram showing a process for predicting the actual available delivery time and the list of available delivery time periods based on the cargo owner's historical order delivery data, as proposed in Example 1 of the present invention;

[0065] Figure 3 A schematic diagram showing the structure of the logistics pickup vehicle scheduling planning system proposed in Example 3 of the present invention. DETAILED DESCRIPTION

[0066] The accompanying drawings are for illustrative purposes only and are not to be construed as limiting this patent;

[0067] In order to better illustrate this embodiment, some parts of the drawings may be omitted, enlarged, or reduced, and do not represent the actual size;

[0068] It is understandable to those skilled in the art that descriptions of certain well-known contents may be omitted in the drawings.

[0069] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0070] The description of the positional relationship in the accompanying drawings is only for illustrative purposes and should not be construed as a limitation of this patent;

[0071] Embodiment 1

[0072] As Figure 1 shown, this embodiment proposes a scheduling and planning method for logistics pick-up vehicles. Refer to Figure 1 , the method includes:

[0073] S1. Divide the logistics virtual division areas and allocate self-owned pick-up vehicles to each logistics virtual division area;

[0074] In this embodiment, it is very important to divide the logistics virtual division areas. The logistics virtual division areas are to gather important customers and geographical locations with high order density into different virtual areas within the division scope. The actual reference dimensions and weights need to be set针对性 according to the specific business model. Based on the geographical locations of the shippers' shipments, the logistics virtual division areas are divided. When dividing, the historical order pick-up data of the shippers is obtained, and features are extracted from the historical order pick-up data of the shippers. The features include: the number of active shippers, the number of orders, the weight of the goods, and the order placement cycle. Then, clustering operations are performed in combination with machine learning algorithms, and the geographical locations of each shipper's shipment are divided into c1*w1 logistics virtual division areas, where c1 is the number of self-owned pick-up vehicles allocated to a certain logistics virtual division area, and w1 is a configuration parameter, 0 < w1 <= 1; in this process, if the obtained historical order pick-up data of the shippers is sufficient, clustering operations can be performed in combination with the CNN neural network. If the data volume is insufficient, traditional machine learning algorithms, such as weighted Kmeans++, are used for clustering. When using this algorithm, the number of regions needs to be set, and the data volume of the regions needs to refer to the number of vehicles in the regions. Each region should have at least one vehicle.

[0075] In addition, because the actual geographical area included in the division is large, after the logistics virtual division areas are actually divided, the range with low order density in the edge area is trimmed to control the radius of the area, which can perform scheduling more efficiently. That is, after the logistics virtual division areas are divided, the logistics virtual division areas are trimmed. During actual execution, the radius value of the configured virtual area center can be used, and the area exceeding the configured radius can be separated from the corresponding area. The initial radius value for reference is 5-10 km.

[0076] S2. Estimate the daily order volume of each logistics virtual division area and allocate self-owned pick-up vehicles according to the estimated daily order volume of each logistics virtual division area;

[0077] In step S2, the estimate takes into account the week-on-week changes, holidays, and promotional periods. The daily order volume estimate includes both non-holiday and holiday order volume estimates. The non-holiday order volume y1 is estimated using the following formula:

[0078] y1=(c2*w2+c3*(1-w2))*c4 / c5*w3

[0079] Among them, c2 represents the daily order volume for the same day of the same week last month; c3 represents the daily order volume for the same day of the previous week; the month-on-month change in the daily order volume estimate includes both monthly and weekly changes; w2 represents the proportion of the monthly change in the daily order volume estimate, which is a configured value, 0 <= w2 <= 1; c4 represents the total order volume for the past 7 days; c5 represents the total order volume for the corresponding 7 days of c4 in the same period last month; and w3 represents the influencing factor of the promotion cycle.

[0080] The estimated formula for holiday order quantity y2 is:

[0081] y2=c6 / c7*c8*(1+w4)

[0082] Among them, c6 represents the total daily order volume in the 7 days before the statutory holiday in this cycle; c7 represents the total daily order volume in the 7 days before the same statutory holiday in the previous cycle; c8 represents the daily order volume on the same statutory holiday in the previous cycle; w4 represents the holiday price adjustment coefficient, -1 <w4<1;

[0083] The formula for allocating self-owned pickup vehicles based on the estimated daily order volume of each virtual logistics branch area is:

[0084]

[0085] Among them, x p represents the self-owned pickup vehicle allocated to the pth logistics virtual branch area, y i p It represents the estimated daily order volume of the p-th logistics virtual branch area. On non-holidays, i is 1, and on holidays, i is 2. w6 represents the total number of self-owned pickup vehicles configured.

[0086] In actual implementation, historical order data can be obtained and calculated in the early morning of the same day. For example, at 2 a.m., w2, w3, and w4 can be configured as follows: w2 is 0.5, w3 is 0.3, and w5 is -0.5. Enterprises can adjust the configuration parameters according to the actual order data of the business. For example, during the peak season, the order volume grows rapidly. Based on the order growth trend, w2 can be set to 0.3 to give a higher weight to recent sales.

[0087] Based on the estimated number of orders in each virtual logistics branch area, the branch's vehicles need to be allocated to different virtual logistics branch areas in proportion. Each area needs to be guaranteed to have at least one vehicle allocated to it. What is set here is the proportion of vehicles, and no specific vehicles are actually bound. Vehicles are only bound to the corresponding areas when they are actually picked up and dispatched on a daily basis.

[0088] In addition, the actual number of vehicles going to work each day is different due to reasons such as adjustments and holidays. The specific number of vehicles going to work can be manually scheduled. A more flexible way is to use the vehicle positioning system to determine the daily active vehicles and confirm the daily vehicle information.

[0089] S3. Predict the actual pickup time and available pickup time slots based on the shipper's historical order pickup data;

[0090] In the field of large-ticket LTL, small and medium-sized enterprise shippers ship their cargo according to their actual production situation. The actual delivery time will be adjusted according to the production plan, and predictions must be made based on different types of shippers. Therefore, in step S3, feature elements are obtained from the shipper's historical order pickup data. Feature elements include: shipper's order time, expected pickup time, and actual pickup time. Figure 2 The process of predicting the actual available pickup time and the list of available pickup time periods based on the shipper's historical order pickup data includes:

[0091] S31. Calculate the mean square error between the actual pickup time and the expected pickup time and perform normalization.

[0092] Among them, the calculation formula of mean square error f is:

[0093]

[0094] in, The mean of the difference between the actual delivery time and the expected delivery time of the consignor, d p It represents the difference between the actual pickup time and the expected pickup time for a single order on a specific day of the week. In this case, we first count the number of orders placed by each shipper on each day of the week, the average delay between the expected pickup time and the actual pickup time, the total number of orders placed, the total expected pickup time and the average delay between the actual pickup time, and then put them into the formula for calculation and save the data to the database. The normalized formula is:

[0095]

[0096] Among them, f min represents the minimum value of the mean square error, f mix represents the maximum value of the mean square error, and g represents the value after normalization;

[0097] S32. Predict the delay time t of the actual delivery time of the consignor, expressed as:

[0098]

[0099] Among them, r is the actual delivery time of the order corresponding to the historical Sunday of the consignor's historical order, h is the expected delivery time of the order corresponding to the historical Sunday of the consignor's historical order, n is the total number of orders, j is a configuration parameter, and the default value is 2;

[0100] S33. Add the expected delivery time of the order on that day to the delay time t of the actual delivery time of the consignor to obtain the predicted actual delivery time;

[0101] S34. A list of available pickup time periods is generated based on the actual pickup time data of the consignor's historical orders. If the consignor has a history of picking up goods during a corresponding time period, the corresponding time period is added to the list of available pickup time periods. This is used to select the next available pickup time period if the predicted time period cannot be met. If the time efficiency service cannot be met, a list of available pickup time periods for the consignor needs to be generated. Here, the consignor's actual pickup time periods are counted to determine the consignor's available pickup time periods.

[0102] A better method is to better confirm the shipper's available pickup time by the number of actual pickup times and their proportion.

[0103] S4. Shipowner Order: Based on the number of days the shipper places an order within a week, combined with the actual available pickup time and available pickup time period predicted by S3, the shipper's actual available pickup time and available pickup time period are predicted. Combined with S1, the shipper's virtual logistics branch area is determined.

[0104] When the consignor plans to ship the goods and opens the consignor's order client to place an order, after the back-end system confirms the order, it will generate the current consignor's actual available pickup time and available pickup time list based on the consignor's expected pickup time, the consignor's pickup delay and the consignor's available pickup time confirmed in step S3.

[0105] For new shippers, in the absence of historical order data, it is necessary to use the historical average data of similar new shippers to make a forecast. Similarly, for new shippers, a default list of available pickup hours should be set, such as: 10, 11, 14, 15, 17, 18, etc. as general pickup hours.

[0106] Because vehicles are dynamically bound every day, when the current area has not yet bound a vehicle, it is necessary to confirm the number of vehicles on duty that day and select nearby vehicles that are not bound by other areas according to the vehicle ratio planned in S1. Specifically, execute step S5.

[0107] S5. Determine whether the cargo pickup vehicle assigned to the cargo owner's virtual logistics branch area can pick up the cargo within the cargo owner's actual available pickup time and available pickup time slot list. If so, dispatch the cargo pickup vehicle. Otherwise, search for a nearby vehicle within the assigned range for pickup.

[0108] When judging whether the self-owned pickup vehicles equipped in the logistics virtual branch area to which the shipper belongs can pick up the goods within the current shipper's actual pickup time and pickup time period list, determine the self-owned pickup vehicles allocated to the logistics virtual branch area according to the logistics virtual branch area to which the shipper belongs, and check whether there is a vehicle that can pick up the goods in the corresponding actual pickup time and pickup time period list. If there is a self-owned pickup vehicle that can pick up the goods within the current shipper's actual pickup time and pickup time period list, dispatch the self-owned pickup vehicle to pick up the goods, and create a vehicle pickup record, including order information, pickup time period and vehicle. First, obtain the list of vehicles bound to the current area and check the scheduled pickup plan for each vehicle.

[0109] If the current vehicle is free during the predicted pickup time period, or there is pickup available during the corresponding time period, but the driving distance between the two areas is close (closer is the parameter configuration value, the actual configuration can be 1KM), then this vehicle can be arranged to pick up the goods. If no regional vehicle can pick up the goods during the corresponding time period, then according to the consignor's list of available pickup time periods, the next time period of the consignor's available pickup time period will be matched, and the matching process will be repeated until a pickup vehicle is matched or all the consignor's available pickup time periods are matched.

[0110] In addition, in the field of large-ticket LTL, since the cargo of shippers is generally heavy and loading takes a long time, in order to more reasonably assess whether the vehicle can pick up the cargo, the parameters can be set to configure the loading delay, with a reference value of 20 minutes;

[0111] If there is no self-owned pickup vehicle that can pick up the goods within the current shipper's actual available pickup time and available pickup time period list, a neighboring vehicle is searched within the configuration range. The neighboring vehicle refers to a vehicle that is idle during the predicted pickup time period or within the configuration range of the pickup time period and pickup area. The configuration range is a first configuration range. The first configuration range is centered on the center of the virtual logistics sub-area to which the shipper belongs and has a radius of 5km. The neighboring vehicle determines all available pickup time periods of the shipper until an available pickup vehicle and the corresponding available pickup time period are output, or until all neighboring vehicles are unable to pick up the goods, and then step S6 is executed;

[0112] S6. If a neighboring vehicle cannot be scheduled for pickup, the vehicle is matched and evaluated with a pickup vehicle owned by another virtual logistics division. If the match is successful, the vehicle is assigned to the pickup. Otherwise, the cost of renting an external vehicle is compared with the cost of arranging a pickup vehicle owned by another virtual logistics division. The vehicle with the lower cost is selected for pickup.

[0113] The process of matching and evaluating the self-owned pickup vehicles in other virtual logistics division areas is as follows:

[0114] S61. Check whether the owned pickup vehicles in other virtual logistics divisions are available within the current shipper's actual available pickup time and available pickup time period, and within the second configuration range. The locations of the cross-regional idle vehicles corresponding to the second configuration range are determined using the center point of the previous order pickup location and the next order pickup location as reference points, with a radius greater than the radius of the first configuration range (in this embodiment, 10 km). When the cross-regional distance interval is met, determine whether a vehicle can be scheduled for pickup based on the predicted order volume of the virtual region.

[0115] S62. If the average number of pickup orders per vehicle per hour parameter w7 is configured within the second configuration range and is set to 1.5, the following formula is used to evaluate whether the self-owned pickup vehicles in other virtual logistics divisions can be used for pickup:

[0116] rd=(r1-r2) / r3 / (r4-r5)

[0117] Where rd represents the estimated average number of pickup orders per vehicle per hour, r1 represents the estimated number of orders for the day corresponding to other virtual logistics divisions, r2 represents the actual number of orders for the day corresponding to other virtual logistics divisions, r3 represents the number of self-owned pickup vehicles bound to other virtual logistics divisions, r4 represents the order cut-off hours for the day, and r5 represents the actual available pickup hours at that time.

[0118] If rd>=w7, then the self-owned pickup vehicles in other virtual logistics division areas cannot be used to pick up goods. Otherwise, the self-owned pickup vehicles in other virtual logistics division areas can be used to pick up goods.

[0119] S63. If outside the second configuration range, compare the cost of renting an external vehicle with the cost of arranging a self-owned pickup vehicle for another virtual logistics branch area. The cost calculation expression for arranging a self-owned pickup vehicle for another virtual logistics branch area is:

[0120] cost1=k1+k2*2*k3 / 100*L

[0121] Where cost1 represents the cost of arranging self-owned pickup vehicles in other virtual logistics divisions; k1 represents the average fixed cost of a single pickup by a self-owned pickup vehicle; k2 represents the distance traveled for the pickup; k3 represents the average fuel consumption per 100 kilometers of the self-owned pickup vehicle; and L represents the fuel price per liter.

[0122] The cost calculation expression for renting an external vehicle is:

[0123] cost2=k4+(k2*2-k5)*q

[0124] Where cost2 represents the cost of renting an external vehicle; k2 represents the distance traveled to pick up the goods; k4 represents the starting price for renting different grades of vehicles according to the weight of the goods; k5 represents the number of kilometers rented by the external vehicle, and q represents the unit price for renting an external vehicle for additional kilometers.

[0125] For details, see Table 1. Taking cargo weight as an example, see the following configuration:

[0126] Table 1

[0127]

[0128] When renting external vehicles, you can consider integrating with the system of a cooperating fleet to automatically rent vehicles. You can also manually input the corresponding rental vehicle information into the backend system.

[0129] When cost1>=cost2, arrange for self-owned pickup vehicles in other logistics virtual branch areas to pick up the goods; otherwise, rent external vehicles to pick up the goods.

[0130] Example 2

[0131] After step S6 mentioned in Example 1, a pick-up plan for the logistics pickup vehicle is generated. The pick-up plan includes the pick-up time period and the vehicle used for picking up the goods. The dispatcher notifies the consignor via SMS and notifies the consignor to reserve the pick-up time period and the vehicle used for picking up the goods according to the pick-up plan. When the consignor changes the pick-up time period, the dispatcher changes the pick-up time period to the vehicle's idle time period. The idle time period of the self-owned pickup vehicle is extracted from the scheduling plan of the self-owned pickup vehicle. The idle time period of the external vehicle is reassigned once. The stage of changing the pick-up time period is when the consignor finally confirms the pick-up time period. The corresponding driver of the self-owned pickup vehicle is notified in real time through the mobile terminal system, and the driver of the rented external vehicle is manually notified by the dispatcher. In addition, for the self-owned pickup vehicle, the driving route and expected time can be provided through the map system to assist the driver in better setting off to pick up the goods.

[0132] According to the pick-up plan, the mobile client reminds the driver to pick up the goods, and the driver picks up the goods according to the agreed time period.

[0133] After the goods are loaded onto the pickup vehicle, the driver of the vehicle clicks the pickup completion button through the mobile client. For externally rented vehicles, the dispatcher manually confirms and completes the pickup through the PC management background operation.

[0134] In addition, the entire logistics system can also locate the pickup vehicle, confirm the time when the pickup vehicle arrives and leaves the pickup location area, and automatically calculate the approximate time point when the pickup is completed.

[0135] Among them, two parameters need to be configured:

[0136] The radius of the pickup area. This parameter needs to be adjusted based on the specific environment and equipment accuracy. The reference setting value is 200.

[0137] The length of time for confirming the pickup area is used to confirm the length of time after the pickup vehicle enters the pickup area. It is used to determine the pickup vehicle's progress in the pickup process and to reduce misjudgment of the corresponding area. The reference setting value is 5 minutes.

[0138] After cargo is picked up and loaded, the vehicles are shunted. After shunting is complete, the owned pickup vehicles return to their respective virtual logistics divisions, awaiting the next pickup schedule. It should be noted that in this embodiment, the logistics system automatically dispatches drivers for shunting in order to deliver cargo to trunk lines as quickly as possible and repurpose the vehicles for pickup, thereby increasing the utilization rate of owned pickup vehicles.

[0139] Example 3

[0140] like Figure 3 As shown, this embodiment also proposes a logistics delivery vehicle scheduling planning system, see Figure 3 , the system comprises:

[0141] The virtual division configuration module 101 is used to divide the logistics virtual division area and equip each logistics virtual division area with its own delivery vehicle;

[0142] The order volume estimation configuration module 102 is used to estimate the daily order volume of each virtual logistics division area and allocate self-owned delivery vehicles based on the estimated daily order volume of each virtual logistics division area;

[0143] The delivery prediction module 103 predicts the actual delivery time and delivery period list based on the delivery data of the shipper's historical orders;

[0144] The consignor module 104 is used for consignor orders: based on the number of days the consignor places an order within a week, combined with the actual available pickup time and available pickup time period predicted by the pickup prediction module 103, the current consignor's actual available pickup time and available pickup time period are predicted, and combined with the virtual branch division configuration module 101, the virtual logistics branch area to which the consignor belongs is determined;

[0145] The judgment module 105 is used to determine whether the self-owned pickup vehicle allocated in the logistics virtual branch area to which the cargo owner belongs can pick up the cargo within the current cargo owner's actual available pickup time and available pickup time period list. If so, the self-owned pickup vehicle is dispatched to pick up the cargo; otherwise, a nearby vehicle is searched within the allocated range for pickup dispatch;

[0146] The matching evaluation module 106 matches and evaluates the self-owned pickup vehicles in other virtual logistics division areas when the neighboring vehicles cannot be scheduled for pickup. If the matching evaluation is passed, the self-owned pickup vehicle is arranged for pickup. Otherwise, the cost of renting an external vehicle is compared with the cost of arranging the self-owned pickup vehicles in other virtual logistics division areas, and the vehicle with the lower cost is selected as the vehicle for pickup.

[0147] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all embodiments here. 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 claims of the present invention.

Claims

1. A method for scheduling and planning logistics delivery vehicles, characterized in that: The method includes: S1. Divide the logistics virtual division areas and allocate self-owned pick-up vehicles to each logistics virtual division area; S2. Estimate the daily order volume of each logistics virtual division area, and allocate self-owned pick-up vehicles according to the estimated daily order volume of each logistics virtual division area; S3. Predict the actual pick-up time and the list of available pick-up time periods based on the historical order pick-up data of the consignor; S4. The consignor places an order: According to the number of days the consignor places an order within a week, combine the actual pick-up time and the list of available pick-up time periods predicted in S3, predict the actual pick-up time and the list of available pick-up time periods of the current consignor, and combine with S1 to determine the logistics virtual division area to which the consignor belongs; S5. Determine whether the self-owned pick-up vehicles allocated within the logistics virtual division area to which the consignor belongs can pick up the goods within the actual pick-up time and the list of available pick-up time periods of the current consignor. If so, dispatch the self-owned pick-up vehicles to pick up the goods; otherwise, search for neighboring vehicles within the configuration range for pick-up dispatching, and execute step S6; S6. If neighboring vehicles cannot be dispatched for pick-up, match and evaluate the self-owned pick-up vehicles of other logistics virtual division areas. If the matching and evaluation pass, arrange for the self-owned pick-up vehicle to be dispatched for pick-up; otherwise, compare the cost of renting external vehicles with the cost of arranging the self-owned pick-up vehicles of other logistics virtual division areas, and select the one with the lower cost as the vehicle for pick-up to pick up the goods.

2. The method for scheduling and planning logistics delivery vehicles according to claim 1, characterized in that: In step S1, divide the logistics virtual division areas based on the geographical location of the consignor's goods shipment. When dividing, obtain the historical order pick-up data of the consignor, extract features from the historical order pick-up data of the consignor. The features include: the number of active consignors, the number of orders, the weight of the goods, and the order placement cycle. Then, perform clustering operations in combination with machine learning algorithms, and divide the geographical location of each consignor's goods shipment into c1*w1 logistics virtual division areas, where c1 is the number of self-owned pick-up vehicles allocated to a certain logistics virtual division area, w1 is a configuration parameter, and 0 < w1 <= 1; after completing the division of the logistics virtual division areas, trim the logistics virtual division areas for efficient pick-up dispatching.

3. The method for scheduling and planning logistics delivery vehicles according to claim 2, characterized in that: In step S2, when estimating, consider the month-on-month ratio of the number of days in a week, holidays, and promotion cycles. The daily order volume estimate includes the estimate of the order volume on non-holidays and the estimate of the order volume on holidays. The estimate formula for the order volume y1 on non-holidays is: y1 = (c2*w2 + c3*(1 - w2))*c4 / c5*w3 where c2 represents the daily order volume on the same day of the same week in the previous month; c3 represents the daily order volume on the same day of the previous week; the month-on-month ratio in the daily order volume estimate includes the month-on-month ratio and the week-on-week ratio, w2 represents the proportion of the month-on-month ratio in the daily order volume estimate, which is a configured value, and 0 <= w2 <= 1; c4 represents the total order volume in the past 7 days; c5 represents the total order volume in the corresponding 7 days in the previous month for c4; w3 represents the influence factor of the promotion cycle; The estimate formula for the order volume y2 on holidays is: y2 = c6 / c7*c8*(1 + w4) Among them, c6 represents the total daily order volume in the 7 days before the statutory holiday in this cycle; c7 represents the total daily order volume in the 7 days before the same statutory holiday in the previous cycle; c8 represents the daily order volume on the same statutory holiday in the previous cycle; w4 represents the holiday price adjustment coefficient, -1 <w4<1; The formula for allocating self-owned pickup vehicles based on the estimated daily order volume of each virtual logistics branch area is: Among them, x p represents the self-owned pickup vehicle allocated to the pth logistics virtual branch area, y i p It represents the estimated daily order volume of the p-th logistics virtual branch area. On non-holidays, i is 1, and on holidays, i is 2. w6 represents the total number of self-owned pickup vehicles configured.

4. The method for scheduling and planning logistics delivery vehicles according to claim 3, characterized in that: In step S3, characteristic elements are obtained from the consignor's historical order pickup data. The characteristic elements include: the consignor's order time, the expected pickup time, and the actual pickup time. The process of predicting the actual available pickup time and the available pickup time period list based on the consignor's historical order pickup data includes: S31. Calculate the mean square error between the actual pickup time and the expected pickup time and perform normalization. Among them, the calculation formula of mean square error f is: in, The mean of the difference between the actual delivery time and the expected delivery time of the consignor, d p It represents the difference between the actual pickup time and the expected pickup time for a single order on a specific day of the week. The normalized formula is: Among them, f min represents the minimum value of the mean square error, f max represents the maximum value of the mean square error, and g represents the value after normalization; S32. Predict the delay time t of the actual delivery time of the consignor, expressed as: Among them, r is the actual delivery time of the order corresponding to the historical Sunday of the shipper's historical order, h is the expected delivery time of the order corresponding to the historical Sunday of the shipper's historical order, n is the total number of orders, and j is a configuration parameter with a default value of 2; S33. Add the expected delivery time of the order on that day to the delay time t of the actual delivery time of the consignor to obtain the predicted actual delivery time; S34. A list of available pickup time periods is obtained through statistics of the actual pickup time of the shipper's historical orders. If the shipper has a history of picking up goods during the corresponding time period, the corresponding time period is added to the list of available pickup time periods. This is used to select the next available pickup time period when the goods cannot be picked up during the predicted time period.

5. The method for scheduling and planning logistics delivery vehicles according to claim 4, characterized in that: In step S4, when judging whether the self-owned pickup vehicles equipped in the logistics virtual branch area to which the consignor belongs can pick up the goods within the current consignor's actual pickup time and pickup time period list, the self-owned pickup vehicles allocated to the logistics virtual branch area are determined according to the logistics virtual branch area to which the consignor belongs, and whether there is a vehicle that can pick up the goods in the corresponding actual pickup time and pickup time period list is checked. If there is a self-owned pickup vehicle that can pick up the goods within the current consignor's actual pickup time and pickup time period list, the self-owned pickup vehicle is dispatched to pick up the goods, and a vehicle pickup record is created, including order information, pickup time period and vehicle.

6. The method for scheduling and planning logistics delivery vehicles according to claim 5, characterized in that: If there is no self-owned pickup vehicle that can pick up the goods within the current shipper's actual pickup time and pickup period list, the neighboring vehicle is searched within the configuration range. The neighboring vehicle refers to a vehicle that is idle during the predicted pickup period, or is within the pickup period and pickup area configuration range. The configuration range is the first configuration range. The first configuration range is centered on the center of the logistics virtual branch area to which the shipper belongs, and has a radius of 5km.

7. The method for scheduling and planning logistics delivery vehicles according to claim 6, characterized in that: In step S6, the process of matching and evaluating the self-owned pickup vehicles in other logistics virtual division areas is as follows: S61. Check if the owned pickup vehicles in other virtual logistics divisions are available within the current shipper's actual available pickup time and available pickup time slots, and are within a second configuration range. The second configuration range is based on the center of the previous order pickup location and the next order pickup location as reference points, and has a radius greater than the radius of the first configuration range. S62. If the average number of pickup orders per vehicle per hour parameter w7 is configured within the second configuration range, use the following formula to evaluate whether the self-owned pickup vehicles in other logistics virtual division areas can be used for pickup. The formula is: rd＝(r1-r2) / r3 / (r4-r5) Where rd represents the estimated average number of pickup orders per vehicle per hour, r1 represents the estimated number of orders for the day corresponding to other virtual logistics divisions, r2 represents the actual number of orders for the day corresponding to other virtual logistics divisions, r3 represents the number of self-owned pickup vehicles bound to other virtual logistics divisions, r4 represents the order cut-off hours for the day, and r5 represents the actual available pickup hours at that time. If rd>=w7, then the self-owned pickup vehicles in other virtual logistics division areas cannot be used to pick up goods. Otherwise, the self-owned pickup vehicles in other virtual logistics division areas can be used to pick up goods. S63. If outside the second configuration range, compare the cost of renting an external vehicle with the cost of arranging a self-owned pickup vehicle for another virtual logistics branch area. The cost calculation expression for arranging a self-owned pickup vehicle for another virtual logistics branch area is: cost1=k1+k2*2*k3 / 100*L Where cost1 represents the cost of arranging self-owned pickup vehicles in other virtual logistics divisions; k1 represents the average fixed cost of a single pickup by a self-owned pickup vehicle; k2 represents the distance traveled for the pickup; k3 represents the average fuel consumption per 100 kilometers of the self-owned pickup vehicle; and L represents the fuel price per liter. The cost calculation expression for renting an external vehicle is: cost2=k4+(k2*2-k5)*q Where cost2 represents the cost of renting an external vehicle; k2 represents the distance traveled to pick up the goods; k4 represents the starting price for renting different grades of vehicles according to the weight of the goods; k5 represents the number of free kilometers rented by the external vehicle, and q represents the unit price for renting an external vehicle for additional kilometers. When cost1>=cost2, arrange for self-owned pickup vehicles in other logistics virtual branch areas to pick up the goods; otherwise, rent external vehicles to pick up the goods.

8. The method for scheduling and planning logistics delivery vehicles according to claim 7, characterized in that: After step S6, a delivery plan for the logistics delivery vehicle is generated. The delivery plan includes the delivery time period and the vehicle used for delivery. The dispatcher notifies the cargo owner via SMS and informs the cargo owner to reserve the delivery time period and the vehicle used for delivery according to the delivery plan. When the cargo owner changes the pick-up time, the dispatcher changes the pick-up time to the vehicle's idle time that can be reserved. The idle time that can be reserved of the own pickup vehicle is extracted from the scheduling plan of the own pickup vehicle, and the idle time that can be reserved of the external vehicle is reassigned once. The stage of changing the pick-up time is when the cargo owner finally confirms the pick-up time.

9. The method for scheduling and planning logistics delivery vehicles according to claim 8, characterized in that: After picking up the goods, the cargo loading is completed and the vehicles are diverted. After the diversion is completed, the owned pickup vehicles return to their respective logistics virtual branch areas and wait for the next pickup dispatch.

10. A logistics delivery vehicle scheduling and planning system, characterized in that: include: Virtual division configuration module, used to divide logistics virtual division areas and equip each logistics virtual division area with its own delivery vehicles; The order volume estimation configuration module is used to estimate the daily order volume of each virtual logistics branch area and allocate self-owned pickup vehicles based on the estimated daily order volume of each virtual logistics branch area; The delivery forecast module predicts the actual delivery time and delivery period based on the cargo owner's historical order delivery data; The consignor module is used for consignor orders: Based on the number of days the consignor places an order within a week, combined with the actual available pickup time and available pickup time period predicted by the pickup prediction module, the current consignor's actual available pickup time and available pickup time period are predicted, and combined with the virtual division configuration module, the consignor's logistics virtual division area is determined; The judgment module is used to determine whether the self-owned pickup vehicles deployed in the virtual logistics branch area to which the cargo owner belongs can pick up the cargo within the current cargo owner's actual available pickup time and available pickup time list. If so, the self-owned pickup vehicle is dispatched to pick up the cargo; otherwise, a nearby vehicle is searched within the configured range for pickup scheduling; The matching and evaluation module matches and evaluates the self-owned pickup vehicles in other virtual logistics division areas when neighboring vehicles cannot be scheduled for pickup. If the matching evaluation passes, the self-owned pickup vehicle is arranged for pickup. Otherwise, the cost of renting an external vehicle is compared with the cost of arranging the self-owned pickup vehicles in other virtual logistics division areas, and the vehicle with the lower cost is selected as the vehicle for pickup.

Images