Internet-based canteen delivery path planning method and system

Abstract

The application relates to the technical field of path optimization, in particular to a canteen meal delivery path planning method and system based on the Internet, which comprises the following steps: based on the congestion interference caused by the overlapping of busy periods between adjacent canteen stalls, combining the congestion performance of the canteen stalls, determining the local congestion comprehensive influence of each busy period of the canteen stalls, based on the meal taking time of the meal delivery personnel, determining the comprehensive meal taking delay condition and secondary order taking risk degree of each busy period of the canteen stalls, comprehensively considering the comprehensive meal taking delay condition and secondary order taking risk degree of each meal taking stall corresponding busy period in each selected meal delivery path, and the local congestion comprehensive influence of each non-meal taking stall corresponding busy period, obtaining the interference degree of each selected meal delivery path, and screening the final meal delivery path according to the interference degree; when the meal is delivered by the final meal delivery path, the meal delivery efficiency can be improved, the student meal experience can be ensured, and the service quality can be significantly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path optimization, in particular to an Internet-based canteen delivery path planning method and system. BACKGROUND

[0002] With the development of big data, the Internet, cloud computing and other technologies, many university canteens have begun to use intelligent delivery systems based on Internet technology. These systems can track students' ordering needs in real time, analyze delivery needs in different areas, and optimize delivery paths based on real-time traffic conditions and weather factors. Various advanced technologies such as big data, artificial intelligence, unmanned delivery and intelligent scheduling systems are providing efficient, energy-saving and intelligent solutions for school canteen delivery. These technologies not only improve delivery efficiency, but also reduce costs and improve students' dining experience.

[0003] Current canteen delivery path planning is affected by many dynamic factors, especially real-time road congestion changes caused by factors such as window queue numbers and student meal peak periods, making traditional path planning algorithms unable to effectively respond to these dynamic changes. In particular, when multiple orders are delivered simultaneously, and there is congestion at various stalls in the canteen, the intersection and overlap of different routes, as well as stall congestion, all exacerbate the complexity of the delivery process, leading to the deterioration of dynamic congestion. This not only increases the difficulty of path planning, but also further affects delivery efficiency and students' dining experience, especially during meal peak periods, causing students to wait for a long time and delivery delays, significantly reducing service quality. SUMMARY

[0004] In order to solve the technical problem that the existing canteen delivery path planning method affects delivery efficiency, the purpose of the present application is to provide an Internet-based canteen delivery path planning method and system, and the technical solution adopted is as follows:

[0005] In the first aspect of the present application, an Internet-based canteen delivery path planning method is provided, comprising:

[0006] Based on the congestion interference caused by the overlap of busy periods between adjacent canteen stalls, combined with the congestion performance of the canteen stalls, the local congestion comprehensive influence of each busy period of the canteen stalls is determined, and the congestion performance is related to the passenger flow concentration and the meal serving speed of the canteen stalls;

[0007] Based on the meal taking time of the delivery personnel, the comprehensive meal taking delay of each busy period of the canteen stalls and the secondary order taking risk degree are determined, and the secondary order taking risk degree represents the possibility of secondary meal taking by the delivery personnel;

[0008] determining each candidate delivery path of the delivery personnel, the candidate delivery path including a pickup gate and a non-pickup gate;

[0009] obtaining an interference degree of each candidate delivery path by synthesizing a comprehensive pickup delay condition of each pickup gate in each candidate delivery path corresponding to a busy period and a secondary order risk degree and a local congestion comprehensive influence of each non-pickup gate corresponding to the busy period, the interference degree being used to indicate a final delivery path screening.

[0010] In an exemplary embodiment, the congestion performance acquisition process comprises:

[0011] acquiring a daily passenger flow concentration condition, a dish serving time consumption condition and a total order quantity of the canteen gate;

[0012] synthesizing the daily passenger flow concentration condition and the dish serving time consumption condition by taking the total order quantity as a fusion adjustment reference to obtain a daily congestion performance component of the canteen gate; the total order quantity positively adjusts the passenger flow concentration condition;

[0013] synthesizing the daily congestion performance component of the canteen gate to obtain the congestion performance of the canteen gate.

[0014] In an exemplary embodiment, the passenger flow concentration condition acquisition process comprises:

[0015] acquiring an order quantity of each unit time period of the canteen gate each day and obtaining a maximum order quantity therefrom;

[0016] acquiring an average value of a difference between the maximum order quantity and an order quantity in each unit time period of the same day;

[0017] obtaining a daily passenger flow concentration condition of the canteen gate according to the average value.

[0018] In an exemplary embodiment, the local congestion comprehensive influence acquisition process comprises:

[0019] determining a busy performance of each busy period of the canteen gate, the busy performance being related to an order quantity of the busy period;

[0020] determining an overlap condition of a candidate busy period of a current gate and a busy time total interval of adjacent gates of the current gate; the current gate is any one of the canteen gates, the candidate busy period is any one of the busy periods of the current gate, and the busy time total interval is obtained from all busy periods of the adjacent gates;

[0021] obtaining a local mutual congestion influence of adjacent gates of the candidate busy period according to the overlap condition of the candidate busy period and the busy performance;

[0022] The local congestion comprehensive influence of the candidate busy time period is obtained by fusing the local mutual congestion influence of adjacent stalls in the candidate busy time period and the congestion performance of the current stall.

[0023] In an exemplary embodiment, the busy time period acquisition process comprises:

[0024] The order quantity of each unit time period of the canteen stall is determined.

[0025] The average order quantity of each unit time period is obtained by acquiring the average value of the order quantity of the same unit time period in all days.

[0026] The target unit time period is a unit time period with an average order quantity greater than a preset reference order quantity.

[0027] In an exemplary embodiment, the comprehensive take-out delay condition acquisition process comprises:

[0028] The take-out duration of the delivery personnel in the candidate busy time period of the current stall is determined; the current stall is any canteen stall, and the candidate busy time period is any busy time period of the current stall.

[0029] The comprehensive take-out delay condition of the candidate busy time period is obtained by fusing the take-out duration of the candidate busy time period and the local congestion comprehensive influence.

[0030] In an exemplary embodiment, the secondary order risk degree acquisition process comprises: obtaining the secondary order risk degree of the candidate busy time period according to the take-out duration of the delivery personnel in the candidate busy time period.

[0031] In an exemplary embodiment, the interference degree acquisition process comprises:

[0032] The comprehensive take-out delay condition of the busy time period in which the time at which the delivery personnel arrives at each take-out stall in the candidate delivery path is located, and the secondary order risk degree are obtained, and the take-out delay feature and the secondary order risk feature of the candidate delivery path are obtained respectively.

[0033] The route obstruction degree of the candidate delivery path is determined, and the route obstruction degree is obtained from the local congestion comprehensive influence of the busy time period in which the time at which the delivery personnel arrives at each non-take-out stall in the candidate delivery path is located.

[0034] The take-out delay feature is taken as a fusion adjustment reference to fuse the secondary order risk feature and the route obstruction degree of the candidate delivery path, and the interference degree of the candidate delivery path is obtained; the take-out delay feature positively adjusts the secondary order risk feature.

[0035] In an exemplary embodiment, the Internet-based canteen delivery path planning method further comprises: taking the candidate delivery path corresponding to the minimum interference degree in the interference degrees of the candidate delivery paths as the final delivery path.

[0036] In a second aspect of the present application, an Internet-based canteen delivery path planning system is provided, comprising: a memory and a processor; the memory is connected with the processor; the memory is used for storing program instructions; and the processor is used for implementing the above-mentioned Internet-based canteen delivery path planning method when the program instructions are executed.

[0037] The present application has the following beneficial effects: in the delivery path planning, the congestion performance of the canteen counter itself, the mutual congestion interference caused by the overlapping of the busy time periods between adjacent canteen counters, and the influence of the meal taking delay of each busy time period of the canteen counter and the secondary order taking caused by the meal taking time of the delivery personnel are considered at the same time, so as to determine the final delivery path from the plurality of candidate delivery paths by comprehensively considering various influencing factors, and the delivery efficiency can be improved when the final delivery path is used for delivery, the student meal experience is ensured, and the service quality is significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 is a flowchart of an Internet-based canteen delivery path planning method provided by an embodiment of the present application;

[0039] Figure 2 is a flowchart of obtaining congestion performance provided by an embodiment of the present application;

[0040] Figure 3 is a flowchart of obtaining passenger flow concentration provided by an embodiment of the present application;

[0041] Figure 4 is a flowchart of obtaining busy time period provided by an embodiment of the present application;

[0042] Figure 5 is a flowchart of obtaining local congestion comprehensive influence provided by an embodiment of the present application;

[0043] Figure 6 is a flowchart of obtaining interference degree provided by an embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined inventive objectives, the specific embodiments, structures, features and effects of the present application are described in detail below in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The data information collected in this application is obtained with the full authorization of the relevant parties.

[0046] The embodiment provides an Internet-based canteen delivery path planning method, and the overall idea is: through dynamic analysis of factors such as congestion influence of canteen stalls in each busy period, congestion mutual influence between adjacent stalls, and possible secondary meal taking risks, the interference situation of each candidate delivery path of the delivery personnel is comprehensively determined, and the optimal delivery path is determined, so as to improve the overall delivery efficiency.

[0047] The application scenario of the Internet-based canteen delivery path planning method is: mainly applied to college canteen delivery. There are multiple canteen stalls in the college canteen, and different canteen stalls usually provide different types of catering services. Multiple different meal taking points are set in the school, and each meal taking point is usually distributed in different teaching buildings, dormitory buildings and public areas. The planning of the delivery path needs to be comprehensively analyzed and determined between different canteen stalls and meal taking points, considering dynamic congestion, multi-order cross balance and other factors. It should be understood that, due to the differences in food characteristics of different canteen stalls, the degree of liking of students and teachers, different canteen stalls will show different queuing conditions at meal time. In addition, it should be noted that the Internet-based canteen delivery path planning method is applicable to multiple canteen stalls arranged in sequence in the canteen, and the spacing between adjacent stalls is relatively small, so as to ensure that students can dine; it is not applicable to the following scenarios: a canteen with only one stall, or the spacing between adjacent stalls is too large and does not constitute mutual congestion influence.

[0048] In an exemplary embodiment, the Internet-based canteen delivery path planning method can be configured as an online delivery system. In the online delivery system, customers (i.e. school personnel such as students, teachers, etc.) operate the online delivery system to place orders, determine meal taking points, select canteen stalls and corresponding meals, generate online orders, and delivery personnel take meals from corresponding canteen stalls according to online orders in the online delivery system and deliver them to corresponding meal taking points. It should be understood that the online ordering system can be an APP installed in the user's smart phone for customers to operate.

[0049] As another possibility, customers can also order offline at the canteen counter, thereby generating offline orders. In this case, in addition to the online delivery system, the Internet-based canteen delivery path planning method is also configured as an offline ordering system. Customers operate the offline delivery system at the canteen counter to place orders, or customers inform canteen counter staff at the canteen counter, and canteen counter staff operate the offline ordering system to place orders, thereby generating offline orders. It should be understood that the offline ordering system can be an APP installed in the user's smart phone for customers to operate, or it can be an operating device installed at the canteen counter, which can be operated by both customers and canteen counter staff. It should be understood that the online ordering system and the offline ordering system can be two independent operating systems, such as two independent APPs, or they can be two different functional modules in the same APP.

[0050] The personnel queuing in front of each counter are not only customers waiting to eat, but also delivery personnel waiting to pick up meals. The gathering of personnel in front of each canteen counter will cause local congestion in the local area of the counter, and the duration of this congestion is not only related to the popularity of the counter, but also closely related to the meal delivery speed of the counter (for example, fast food meals, even if the number of people in line is large, the meal delivery speed is fast; for meals that require secondary processing, such as hot pot and casserole, the meal delivery speed is relatively slow). Therefore, for canteen counters, the influence on the congestion in front of the counter can be measured from the popularity and meal delivery speed of the counter.

[0051] Under normal circumstances, there is a certain regularity in canteen meal times, and there are usually two meal peaks in a day, namely the lunch peak and the dinner peak. The time period of the lunch peak is usually 11:30 AM-1:30 PM, and the time period of the dinner peak is usually 5:30 PM-7:30 PM. In order to improve the reliability of the delivery path planning, the present embodiment mainly analyzes the meal peak of each day. Therefore, the daily data information of each canteen counter is the data information of the meal peak of each day. It should be understood that if the data information is specifically the data information of the lunch peak, then the delivery path planning is for the lunch peak delivery path; if the data information is specifically the data information of the dinner peak, then the delivery path planning is for the dinner peak delivery path. The present embodiment takes the lunch peak as an example, and the data information collected is the data information in the time period of the lunch peak.

[0052] The Internet-based canteen delivery path planning method can be configured as a background system that interacts with the online delivery system and the offline ordering system to obtain the online orders and offline orders of each canteen counter each day. Therefore, the orders include online orders and offline orders, and the order volume includes online order volume and offline order volume.

[0053] As Figure 1 shown, the embodiment provides an Internet-based canteen delivery path planning method, comprising:

[0054] Step S1: based on the congestion interference caused by the overlapping of busy periods between adjacent canteen stalls, combined with the congestion performance of the canteen stalls, determine the local congestion comprehensive influence of each busy period of the canteen stalls;

[0055] Step S2: based on the meal taking time of the delivery personnel, determine the comprehensive meal taking delay and secondary order taking risk degree of each busy period of the canteen stalls;

[0056] Step S3: determine the each candidate delivery path of the delivery personnel, the candidate delivery path includes the meal taking stall and the non-meal taking stall;

[0057] Step S4: comprehensively consider the comprehensive meal taking delay and secondary order taking risk degree of each busy period corresponding to each meal taking stall in each candidate delivery path, and the local congestion comprehensive influence of each busy period corresponding to each non-meal taking stall, to obtain the interference degree of each candidate delivery path.

[0058] The specific implementation process of each step will be described in combination with the accompanying drawings.

[0059] Step S1: based on the congestion interference caused by the overlapping of busy periods between adjacent canteen stalls, combined with the congestion performance of the canteen stalls, determine the local congestion comprehensive influence of each busy period of the canteen stalls.

[0060] For multiple stalls in a school canteen, the stalls are usually arranged in sequence, and there are two adjacent stalls according to the arrangement order. In order to facilitate the description, the current stall is set as an arbitrary canteen stall.

[0061] Get the congestion performance of the current stall. The congestion performance is related to the passenger flow concentration of the current stall and the meal serving speed, and the meal serving speed is related to the meal serving time. The more concentrated the passenger flow is, the longer the meal serving time is (i.e. the slower the meal serving speed is), and the more serious the congestion is. In an exemplary embodiment, as Figure 2 shown, a specific process for obtaining the congestion performance is given as follows:

[0062] Step S1-1: get the passenger flow concentration, meal serving time and total order quantity of the canteen stall every day.

[0063] Obtain the customer flow concentration of the current stall every day, the meal serving time and the total order quantity. For the convenience of description, the embodiment presets a monitoring period, which includes multiple days, and the meal delivery path is planned by analyzing the data of each day. It should be understood that the number of days included in the monitoring period is set according to actual needs, such as half a month. Therefore, the monitoring period is a historical time period, and the current meal delivery path is planned by the data information of each day in the historical time period.

[0064] As shown in Figure 3 , the process of obtaining the customer flow concentration includes:

[0065] Step S1-1-1: Obtain the order quantity of each unit time period in the canteen stall every day, and obtain the maximum order quantity therefrom.

[0066] Divide the time period of the lunch peak period of each day into multiple unit time periods for analysis, wherein the length of the unit time period is set according to actual needs, such as 2 minutes. Then, obtain the order quantity of each unit time period in the current stall every day (the order quantity of the unit time period is the sum of the online order quantity and the offline order quantity of the unit time period according to the analysis above), and obtain the maximum order quantity therefrom as the maximum order quantity of the current stall every day. Then, the order quantity of each unit time period in the current stall every day is divided into the maximum order quantity and the order quantity in other unit time periods of the same day.

[0067] Step S1-1-2: Obtain the average value of the difference between the maximum order quantity and the order quantity in other unit time periods of the same day.

[0068] For any day, calculate the difference between the maximum order quantity of the current stall on that day and the order quantity in other unit time periods of that day, and then calculate the average value of the difference. The average value of the difference reflects the overall difference between the maximum order quantity and other order quantities. The larger the value, the more concentrated the customer flow of that day.

[0069] Step S1-1-3: Obtain the customer flow concentration of the canteen stall every day according to the average value.

[0070] The average value of the difference value corresponding to the current stall on the day can represent the customer flow concentration situation of the current stall on the day. In an exemplary embodiment, the average value of the difference value corresponding to the current stall on the day is normalized, and the result is a quantitative result of the customer flow concentration situation of the current stall on the day, which is defined as a customer flow concentration inflow degree, so as to realize the quantification of the customer flow concentration situation of the current stall on the day. In an exemplary embodiment, the normalization method here can be: obtaining the maximum value and the minimum value in the average value of the difference value corresponding to each day in all stalls in the monitoring period, and then using the maximum value and the minimum value normalization method to normalize the average value of the difference value corresponding to the current stall on the day. Thus, the customer flow concentration inflow degree of the current stall on each day is obtained, and the customer flow concentration inflow degree of each stall on each day is obtained.

[0071] The concentration inflow of the customer flow of the current stall not only can exacerbate the local congestion situation in front of the current stall, but also can cause order accumulation due to excessive order quantity, thereby affecting the out-of-order time consumption situation of the current stall. In an exemplary embodiment, for the out-of-order speed of online orders, the customer operates the order on the online delivery system as the starting time, and the current stall staff operates the out-of-order on the online delivery system as the ending time after the food is ready. The time interval between the starting time and the ending time is the out-of-order time, also known as the out-of-order time consumption, which is used to represent the out-of-order speed. The longer the out-of-order time consumption, the slower the out-of-order speed. For the out-of-order speed of offline orders, the customer or the current stall staff operates the order on the offline delivery system as the starting time, and the current stall staff operates the order on the offline delivery system as the ending time after the food is ready. The time interval between the starting time and the ending time is the out-of-order time, which is used to represent the out-of-order speed. The longer the out-of-order time consumption, the slower the out-of-order speed.

[0072] In an exemplary embodiment, for any day, the out-of-order time consumption of each order (including online orders and offline orders) of the current stall on the day is obtained, and the average value is calculated to obtain the average value of the out-of-order time on the day, as the overall situation of the out-of-order time on the day. Then, the average value of the out-of-order time of the current stall on the day is normalized, and the result is a quantitative result of the out-of-order time of the current stall on the day, which is defined as an out-of-order time performance degree, so as to realize the quantification of the out-of-order time of the current stall on the day. In an exemplary embodiment, the normalization method here can be: obtaining the maximum value and the minimum value in the average value of the out-of-order time of each day in all stalls in the monitoring period, and then using the maximum value and the minimum value normalization method to normalize the average value of the out-of-order time of the current stall on the day. Thus, the out-of-order time performance degree of the current stall on each day is obtained, and the out-of-order time performance degree of each stall on each day is obtained.

[0073] The total order volume during the daily lunch peak period of the current stall is obtained and normalized. The result is the daily total order volume performance index of the current stall, thus quantifying the daily total order volume of the current stall. In an exemplary embodiment, the normalization method here can be: obtaining the maximum and minimum values ​​of the daily total order volume among all stalls within the monitoring period, and then using the maximum and minimum value normalization method to normalize the daily total order volume of the current stall. This yields the daily total order volume performance index for each stall.

[0074] Step S1-2: Using the total order volume as a reference for fusion adjustment, the daily customer flow concentration and food preparation time are fused to obtain the daily congestion performance of the canteen stalls.

[0075] To assess the potential congestion at the current food stall, a comprehensive evaluation is conducted. During peak lunch hours, a higher total order volume leads to greater concern about whether orders are concentrated in specific time periods (due to excessive order concentration, long customer wait times and overburdened staff, potentially resulting in inefficient food preparation). Conversely, a lower total order volume during peak hours focuses on the stall's food preparation time (longer preparation times exacerbate queues even with low order volumes). Therefore, the total order volume is used as a fusion adjustment reference to integrate daily customer flow concentration and food preparation time, resulting in a daily congestion performance component. Total order volume positively adjusts for customer flow concentration and negatively adjusts for food preparation time. In an exemplary embodiment, based on the quantification results of each parameter in step S1-1, a specific quantification method for the congestion performance component is given below:

[0076] ;

[0077] in, This represents the congestion performance component of the current stall on day i. This indicates the performance of the total order volume of the current stall on day i. This represents the concentration of customer inflow at the current stall on day i. This represents the time taken to prepare meals at the current food stall on day i.

[0078] Steps S1-3: Combine the daily congestion data of the canteen stalls to obtain the congestion data of the canteen stalls.

[0079] The average value of the congestion performance components for each day within the monitoring period is calculated to obtain the congestion performance of the current stall. This method is used to obtain the congestion performance of each stall.

[0080] During lunch peak, if the adjacent stalls appear crowded at the same time, the queue area of customers will overlap, resulting in too dense queue area, increasing the waiting time of customers, and even making it difficult for some customers to find a queue position in the specified area, causing further congestion of people flow, which will seriously affect the efficiency of taking and delivering meals. Specifically, the congestion of adjacent stalls will affect the stall delivery speed through four aspects of shared space resource occupation, dynamic line cross interference, equipment use queue, and human collaboration efficiency decline: public meal taking area congestion prolongs customer waiting time, leading to backlog of completed orders of stalls; logistics blockage caused by narrow passageway will delay food supply and product delivery; shared cooking equipment needs to be queued for processing, prolonging the meal preparation period; and the staff's operation continuity is reduced due to avoiding the flow of adjacent stalls. Thus, congestion will form a chain reaction of "space congestion-process lag-order accumulation".

[0081] Obtain the busy time period of the current stall every day. In an exemplary embodiment, as shown in FIG. 2, a specific acquisition process of the busy time period is as follows: Figure 4

[0082] Step S2-1: Determine the order quantity of each unit time period of the canteen stall every day.

[0083] Step S2-2: Obtain the average value of the order quantity of the same unit time period in all days to obtain the average order quantity of each unit time period.

[0084] Step S2-3: Form a busy time period by connecting the target unit time periods, and the target unit time period is the unit time period with an average order quantity greater than a preset reference order quantity.

[0085] Among them, the order quantity of each unit time period of the current stall every day is determined. For a certain unit time period, the order quantity in the unit time period every day in the monitoring period is obtained, and the average value of the order quantity in the unit time period of all days is calculated as the average order quantity of the unit time period for all days. Thus, the average order quantity of each unit time period for all days is obtained.

[0086] ​The greater the average order amount, the more busy the corresponding unit time period is. Therefore, a preset reference order amount is set, which can be a preset fixed value or obtained in the following way: the average value of the order amounts of all stalls in all unit time periods is calculated as the preset reference order amount. The average order amount of each unit time period is compared with the preset reference order amount to obtain the unit time period with an average order amount greater than the preset reference order amount, which is defined as a target unit time period. Thus, multiple target unit time periods in the lunch peak period are obtained in time sequence. If there are several target unit time periods in time sequence, these continuous several target unit time periods are connected together to form a busy period, so as to obtain several busy periods of the current stall. It should be understood that for a single target unit time period, i.e. it has no other target unit time period adjacent in time sequence, the single target unit time period is also regarded as a separate busy period.

[0087] The busy periods between the current stall and the adjacent stalls can overlap, which will cause congestion interference. According to the congestion interference caused and the congestion performance of the current stall, the local congestion comprehensive influence of each busy period of the current stall is determined. In an exemplary embodiment, as shown in Figure 5 A specific process for obtaining the local congestion comprehensive influence is given as follows:

[0088] Step S3-1: Determine the busy performance of each busy period of the canteen stall.

[0089] For ease of illustration, the candidate busy period is set as any one of the busy periods of the current stall. According to the order amount of the candidate busy period of the current stall, the busy performance of the candidate busy period is determined, which represents the busy situation of the candidate busy period. In an exemplary embodiment, the total order amount of the current stall per day is obtained, and then the average value of the total order amount is calculated as the total order amount of the lunch peak period of the current stall. The ratio of the order amount of the candidate busy period to the total order amount of the lunch peak period of the current stall is calculated to obtain the result as the busy performance of the candidate busy period. Thus, the busy performance of each busy period of the current stall is obtained.

[0090] Step S3-2: Determine the overlap between the candidate busy period of the current stall and the busy time interval of the adjacent stall of the current stall.

[0091] The stall adjacent to the current stall is determined as the adjacent stall of the current stall. It should be understood that if the current stall is not at the edge position of the canteen, the current stall has left and right adjacent stalls, and if the current stall is at the edge position of the canteen, the current stall has only a left adjacent stall or a right adjacent stall. Taking the case that the current stall has both a left adjacent stall and a right adjacent stall as an example.

[0092] Obtain each busy period of the left adjacent stall of the current stall, and obtain the total busy time interval of the left adjacent stall according to each busy period of the left adjacent stall. In an exemplary embodiment, obtain the union of the time intervals of each busy period of the left adjacent stall as the total busy time interval of the left adjacent stall. Similarly, obtain the union of the time intervals of each busy period of the right adjacent stall as the total busy time interval of the right adjacent stall.

[0093] Obtain the intersection of the time interval of the candidate busy period of the current stall and the total busy time interval of the left adjacent stall as the overlap time of the candidate busy period of the current stall and the total busy time interval of the left adjacent stall, defined as the first overlap time. Similarly, obtain the intersection of the time interval of the candidate busy period of the current stall and the total busy time interval of the right adjacent stall as the overlap time of the candidate busy period of the current stall and the total busy time interval of the right adjacent stall, defined as the second overlap time.

[0094] Obtain the ratio of the time length of the first overlap time to the time length of the candidate busy period as the overlap degree of the candidate busy period of the current stall and the left adjacent stall, defined as the first overlap degree; and obtain the ratio of the time length of the second overlap time to the time length of the candidate busy period as the overlap degree of the candidate busy period of the current stall and the right adjacent stall, defined as the second overlap degree.

[0095] Step S3-3: Obtain the local mutual congestion influence of the adjacent stalls of the candidate busy period according to the overlap situation and the busy performance of the candidate busy period.

[0096] Fuse the first overlap degree and the busy performance of the candidate busy period to obtain the left adjacent stall local mutual congestion influence of the candidate busy period corresponding to the left adjacent stall. Specifically, calculate the product of the first overlap degree and the busy performance of the candidate busy period as the left adjacent stall local mutual congestion influence of the candidate busy period. Similarly, fuse the second overlap degree and the busy performance of the candidate busy period to obtain the right adjacent stall local mutual congestion influence of the candidate busy period corresponding to the right adjacent stall. Specifically, calculate the product of the second overlap degree and the busy performance of the candidate busy period as the right adjacent stall local mutual congestion influence of the candidate busy period.

[0097] Calculate the sum of the left adjacent stall local mutual congestion influence and the right adjacent stall local mutual congestion influence of the candidate busy period as the local mutual congestion influence of the adjacent stalls of the candidate busy period. It should be understood that if the current stall only has the left adjacent stall, the left adjacent stall local mutual congestion influence of the candidate busy period is taken as the local mutual congestion influence of the adjacent stalls; if the current stall only has the right adjacent stall, the right adjacent stall local mutual congestion influence of the candidate busy period is taken as the local mutual congestion influence of the adjacent stalls.

[0098] In order to facilitate subsequent quantification processing, the maximum-minimum normalization method is used to normalize the adjacent local mutual congestion influence of the candidate busy period of the current stall: that is, the maximum and minimum values of the adjacent local mutual congestion influence of all busy periods of all stalls are obtained, and the maximum-minimum normalization method is used to normalize the adjacent local mutual congestion influence of the candidate busy period of the current stall. The adjacent local mutual congestion influence of each busy period of each stall referred to in the following is the normalized result.

[0099] Step S3-4: fuse the adjacent local mutual congestion influence of the candidate busy period and the congestion performance of the current stall to obtain the local congestion comprehensive influence of the candidate busy period.

[0100] The adjacent local mutual congestion influence of the candidate busy period of the current stall and the congestion performance of the current stall affect the local congestion comprehensive influence of the candidate busy period, therefore, the adjacent local mutual congestion influence of the candidate busy period of the current stall and the congestion performance of the current stall are fused, specifically, the product of the adjacent local mutual congestion influence of the candidate busy period of the current stall and the congestion performance of the current stall is calculated as the local congestion comprehensive influence of the candidate busy period of the current stall. By using the above process, the local congestion comprehensive influence of each busy period of the current stall is obtained, and then the local congestion comprehensive influence of each busy period of each stall is obtained.

[0101] Step S2: determine the comprehensive meal taking delay and secondary order risk of each busy period of the canteen stall based on the meal taking time of the meal delivery personnel.

[0102] The pickup time of each online order in the candidate busy period of the current outlet is obtained. The pickup time is the time interval between the time when the delivery personnel arrives at the current outlet and the time when the delivery personnel takes the meal after operating the online delivery system, and the pickup time of each online order in the candidate busy period of the current outlet is obtained. The average value of the pickup time of each online order in the candidate busy period of the current outlet is calculated, and the average pickup time of the candidate busy period of the current outlet is obtained. Then, the average pickup time of the candidate busy period of the current outlet is normalized to obtain the pickup time reference coefficient of the candidate busy period of the current outlet. In an exemplary embodiment, the normalization method is as follows: the maximum and minimum values of the average pickup time of all busy periods of all outlets are obtained, and the maximum and minimum values are normalized to obtain the average pickup time of the candidate busy period of the current outlet. The pickup time reference coefficient of each busy period of each outlet is obtained by using the above process.

[0103] For the local congestion comprehensive influence of the candidate busy period of the current outlet, it is a potential influence that may hinder the pickup of the delivery personnel; and for the pickup time reference coefficient of the candidate busy period of the current outlet, it is a direct influence that directly reflects the delay of the pickup time of the delivery personnel in the candidate busy period of the current outlet. Therefore, the influence on the delivery personnel can be determined from the potential and direct angles. Accordingly, the pickup time reference coefficient and the local congestion comprehensive influence of the candidate busy period of the current outlet are fused to obtain the comprehensive pickup delay of the candidate busy period of the current outlet. Specifically, the local congestion comprehensive influence of the candidate busy period of the current outlet is taken as the weight, the pickup time reference coefficient of the candidate busy period of the current outlet is taken as the reference value, the pickup time reference coefficient of the candidate busy period of the current outlet is multiplied by the local congestion comprehensive influence of the candidate busy period of the current outlet, and the result is taken as the quantitative result of the comprehensive pickup delay of the candidate busy period of the current outlet, which is defined as the comprehensive pickup delay value. The comprehensive pickup delay value represents the delay of the pickup time of the delivery personnel in the pickup and delivery process when passing through the current outlet due to the mutual congestion of the adjacent outlets.

[0104] For the current stall delivery personnel in the lunch peak period, the order quantity of the current stall will usually increase substantially during the pickup process. At this time, too many offline orders may cause the kitchen to be unable to prepare the meals in time. This situation not only affects the meal delivery speed, but also may cause the delivery personnel to waste more time during the pickup process, and even cause secondary pickup (i.e. waiting for meals, only sending other meals first, and then returning to pick up meals). Therefore, based on the pickup time of the delivery personnel, the secondary order risk degree of the candidate busy period of the current stall is determined, and the secondary order risk degree represents the possibility of secondary pickup of the delivery personnel.

[0105] The secondary order risk degree of the candidate busy period of the current stall is related to the average pickup time of the delivery personnel in the candidate busy period of the current stall. According to the average pickup time of the delivery personnel in the candidate busy period of the current stall, the secondary order risk degree of the candidate busy period of the current stall is obtained. In an exemplary embodiment, a maximum pickup time is preset, and the ratio of the average pickup time of the candidate busy period of the current stall to the preset maximum pickup time is taken as the secondary order risk degree of the candidate busy period of the current stall. The preset maximum pickup time can be a reference quantity set by a person, or can be obtained in the following manner: obtaining the longest pickup time of each delivery personnel at each stall in the monitoring period, and selecting the longest pickup time from these longest pickup times as the preset maximum pickup time.

[0106] Therefore, the longer the average pickup time of the candidate busy period of the current stall, the greater the secondary order risk degree, and the higher the possibility of secondary pickup of the delivery personnel in the candidate busy period of the current stall.

[0107] Step S3: determining each candidate delivery path of the delivery personnel, the candidate delivery path including a pickup stall and a non-pickup stall.

[0108] For the delivery personnel, the online delivery system will usually assign multiple orders according to the location of the stall and the pickup point, and the on-the-way situation of each order. Therefore, while delivering meals, the delivery personnel also needs to pick up new orders at the stall on the way, which will cause the dynamic change of the route. Because the delivery personnel not only needs to consider the pickup time during the pickup process, but also needs to consider the time delay caused by personnel congestion when passing through non-pickup stalls. That is, some non-pickup stalls may cause narrow passages or dense personnel due to customers waiting for service or choosing more meals.

[0109] Based on the navigation system in the online delivery system, according to the current order situation, a plurality of delivery paths are automatically generated for the delivery personnel, which are defined as each candidate delivery path. It should be understood that for any one delivery path, the delivery path includes a pickup outlet and a non-pickup outlet. The pickup outlet means that the delivery personnel needs to pick up from the pickup outlet, and the non-pickup outlet means that the delivery personnel does not pick up from the outlet, and only passes through the non-pickup outlet.

[0110] Step S4: The comprehensive pickup delay condition and the secondary order risk degree of each pickup outlet corresponding to the busy period in each candidate delivery path, and the local congestion comprehensive influence of each non-pickup outlet corresponding to the busy period are integrated to obtain the interference degree of each candidate delivery path.

[0111] For any one candidate delivery path, the comprehensive pickup delay condition and the secondary order risk degree of each pickup outlet corresponding to the busy period in the candidate delivery path, and the local congestion comprehensive influence of each non-pickup outlet corresponding to the busy period are integrated to obtain the interference degree of the candidate delivery path. Finally, based on the interference degree of each candidate delivery path, the final delivery path is selected.

[0112] In an exemplary embodiment, as shown in Figure 6 A specific acquisition process of the interference degree is as follows:

[0113] Step S4-1: The comprehensive pickup delay condition of the busy period when the delivery personnel arrives at each pickup outlet in the candidate delivery path, and the secondary order risk degree are obtained, and the pickup delay feature and the secondary order risk feature of the candidate delivery path are obtained respectively.

[0114] For any one pickup outlet, based on the candidate delivery path, the time when the delivery personnel arrives at the pickup outlet is obtained. In an exemplary embodiment, the online delivery system has a positioning function and can obtain the position of the delivery personnel in real time. Combined with the position of the pickup outlet, it is determined when the delivery personnel arrives at the pickup outlet. As another implementation, when the delivery personnel arrives at the pickup outlet, the online delivery system is operated to determine that the store has been arrived, and then the online delivery system obtains the time when the delivery personnel arrives at the pickup outlet. The busy period of the pickup outlet where the delivery personnel arrives at the pickup outlet is obtained. Thus, the comprehensive pickup delay value corresponding to the busy period of the pickup outlet is determined. According to the above process, the time when the delivery personnel arrives at each pickup outlet and the comprehensive pickup delay value of the busy period of each pickup outlet are obtained. Then, the average value of the comprehensive pickup delay value corresponding to each pickup outlet is calculated as the pickup delay feature of the candidate delivery path.

[0115] obtaining a busy period of the pickup counter at the time when the delivery personnel arrives at the pickup counter, so as to determine the secondary order taking risk degree corresponding to the busy period of the pickup counter. According to the above process, the time when the delivery personnel arrives at each pickup counter and the secondary order taking risk degree of the busy period of each pickup counter are obtained. Then, the average of the secondary order taking risk degrees corresponding to each pickup counter is calculated as the secondary order taking risk feature of the candidate delivery path.

[0116] Step S4-2: determining the path blocking degree of the candidate delivery path.

[0117] For any non-pickup counter in any candidate delivery path, a busy period of the non-pickup counter at the time when the delivery personnel arrives at the non-pickup counter is obtained, so as to obtain the local congestion comprehensive influence corresponding to the busy period of the non-pickup counter at the time when the delivery personnel arrives at the non-pickup counter. In this way, the local congestion comprehensive influence of the busy period at the time when the delivery personnel arrives at each non-pickup counter in the candidate delivery path is obtained as the local congestion comprehensive influence of each non-pickup counter in the candidate delivery path. Finally, the average of the local congestion comprehensive influences of all non-pickup counters in the candidate delivery path is calculated as the path blocking degree of the candidate delivery path.

[0118] Step S4-3: taking the order taking delay feature as a fusion adjustment reference to fuse the secondary order taking risk feature and the path blocking degree of the candidate delivery path, so as to obtain the interference degree of the candidate delivery path.

[0119] For the candidate delivery path, the higher the order taking delay feature is, the more attention is paid to the possible secondary order taking, that is, the greater the proportion of the secondary order taking risk feature is; the lower the order taking delay feature is, the more attention is paid to the path blocking degree. Therefore, the order taking delay feature is taken as a fusion adjustment reference to fuse the secondary order taking risk feature and the path blocking degree of the candidate delivery path, so as to obtain the interference degree of the candidate delivery path. The order taking delay feature positively adjusts the secondary order taking risk feature and reversely adjusts the path blocking degree. In an exemplary embodiment, a specific quantification method of the interference degree is as follows:

[0120] ;

[0121] Wherein, L represents the interference degree of the candidate delivery path; Y represents the order taking delay feature of the candidate delivery path; F represents the secondary order taking risk feature of the candidate delivery path; and S represents the path blocking degree of the candidate delivery path.

[0122] In the above manner, the interference degrees of each candidate delivery path are obtained. The smaller the interference degree is, the smaller the interference of external factors on the corresponding candidate delivery path is, and the higher the delivery efficiency is. Therefore, the minimum interference degree is determined from the interference degrees of each candidate delivery path, and the candidate delivery path corresponding to the minimum interference degree is taken as the final delivery path. The delivery personnel performs delivery according to the final delivery path during the lunch peak period, which can ensure that the delivery personnel can more efficiently complete the delivery task.

[0123] It should be understood that for the dinner peak period, the data information in the dinner peak period is obtained, and the data processing process provided in the embodiment is used to obtain the final delivery path of the dinner peak period.

[0124] The embodiment also provides an Internet-based dining hall delivery path planning system, which comprises a memory and a processor, the memory is connected with the processor, the memory is used for storing program instructions, and the processor is used for implementing the steps in the above-mentioned Internet-based dining hall delivery path planning method embodiment when the program instructions are executed.

[0125] In one exemplary embodiment, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps in the above-mentioned Internet-based dining hall delivery path planning method embodiment.

[0126] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0127] Each embodiment in the specification is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other. Each embodiment mainly describes the differences from other embodiments.

Claims

1. An Internet-based canteen delivery route planning method, characterized by, The method comprises the following steps: Based on the congestion interference caused by the overlap of busy periods between adjacent canteen stalls, combined with the congestion performance of the canteen stalls, the local congestion comprehensive influence of each busy period of the canteen stalls is determined, and the congestion performance is related to the passenger flow concentration and the meal serving speed of the canteen stalls; Based on the meal taking time of the delivery personnel, the comprehensive meal taking delay of each busy period of the canteen stalls and the secondary order taking risk degree are determined, and the secondary order taking risk degree represents the possibility of secondary meal taking of the delivery personnel; Determine the each candidate delivery path of the delivery personnel, the candidate delivery path includes the meal taking stall and the non-meal taking stall; Integrate the comprehensive meal taking delay and the secondary order taking risk degree of each busy period corresponding to each meal taking stall in each candidate delivery path, and the local congestion comprehensive influence of each busy period corresponding to each non-meal taking stall, to obtain the interference degree of each candidate delivery path, and the interference degree is used to indicate the final delivery path screening; The congestion performance acquisition process comprises the following steps: acquiring the passenger flow concentration, meal serving time and total order quantity of the canteen stall during lunch peak period every day; the total order quantity is taken as a fusion adjustment reference to fuse the passenger flow concentration and meal serving time during lunch peak period every day to obtain the congestion performance component of the canteen stall during lunch peak period every day; the total order quantity positively adjusts the passenger flow concentration; and the congestion performance components of the canteen stall during lunch peak period every day are fused to obtain the congestion performance of the canteen stall; The specific quantification method of the congestion performance component is as follows: wherein, represents the congestion performance component of lunch peak period of the current stall on the i-th day; represents the total order quantity performance degree of lunch peak period of the current stall on the i-th day, wherein the maximum value and the minimum value of the total order quantity of lunch peak period of each day in all stalls in the monitoring period are obtained, the total order quantity of lunch peak period of each day of the current stall is normalized by using the maximum value and minimum value normalization method, and the total order quantity performance degree of each day of the current stall is obtained; represents the customer flow concentrated inflow degree of lunch peak period of the current stall on the i-th day, wherein the difference between the maximum order quantity of lunch peak period of the current stall on the day and the order quantity in each unit time period in lunch peak period of the day is calculated, the average value of the difference is normalized, and the customer flow concentrated inflow degree of lunch peak period of the current stall on the day is obtained; represents the dish delivery time performance degree of lunch peak period of the current stall on the i-th day, wherein the maximum value and the minimum value of the average value of dish delivery time of lunch peak period of each day in all stalls in the monitoring period are obtained, the average value of dish delivery time of lunch peak period of the day of the current stall is normalized by using the maximum value and minimum value normalization method, and the dish delivery time performance degree of lunch peak period of each day of the current stall is obtained. The local congestion comprehensive influence acquisition process comprises the following steps: determining the busy performance of each busy period of the canteen stall, the busy performance is related to the order quantity of the busy period; determining the overlap of the candidate busy period of the current stall and the busy time total interval of the adjacent stalls; the current stall is any canteen stall, the candidate busy period is any busy period of the current stall, and the busy time total interval is obtained from all busy periods of the adjacent stalls; according to the overlap of the candidate busy period and the busy performance, the local mutual congestion influence of the adjacent stalls of the candidate busy period is obtained; the local mutual congestion influence of the adjacent stalls of the candidate busy period and the congestion performance of the current stall are fused to obtain the local congestion comprehensive influence of the candidate busy period; The interference degree acquisition process comprises the following steps: acquiring the comprehensive meal taking delay of the busy period in which the delivery personnel arrives at each meal taking stall in the candidate delivery path, and the secondary order taking risk degree, respectively obtaining the meal taking delay feature and the secondary order taking risk feature of the candidate delivery path; determining the path obstruction degree of the candidate delivery path, and the path obstruction degree is obtained from the local congestion comprehensive influence of the busy period in which the delivery personnel arrives at each non-meal taking stall in the candidate delivery path; the meal taking delay feature is taken as a fusion adjustment reference to fuse the secondary order taking risk feature and the path obstruction degree of the candidate delivery path to obtain the interference degree of the candidate delivery path; the meal taking delay feature positively adjusts the secondary order taking risk feature; The specific quantification method of the interference degree is as follows: L represents the degree of interference of the candidate delivery path; Y represents the pickup delay characteristic of the candidate delivery path, wherein the local congestion comprehensive influence of the candidate busy period of the current stall is taken as a weight, the pickup time length reference coefficient of the candidate busy period of the current stall is taken as a reference value, the pickup time length reference coefficient of the candidate busy period of the current stall is multiplied by the local congestion comprehensive influence of the candidate busy period of the current stall to obtain a comprehensive pickup delay value, and the average value of the comprehensive pickup delay value corresponding to each pickup stall is calculated as the pickup delay characteristic of the candidate delivery path; F represents the secondary order risk characteristic of the candidate delivery path, wherein a preset maximum pickup time length is taken, the ratio of the average pickup time length of the candidate busy period of the current stall to the preset maximum pickup time length is taken as the secondary order risk degree of the candidate busy period of the current stall, and the average value of the secondary order risk degree corresponding to each pickup stall is calculated as the secondary order risk characteristic of the candidate delivery path; S represents the degree of way obstruction of the candidate delivery path, wherein for any non-pickup stall in any candidate delivery path, the busy period of the non-pickup stall at the time when the delivery personnel arrives at the non-pickup stall is obtained, so as to obtain the local congestion comprehensive influence corresponding to the busy period of the non-pickup stall at the time when the delivery personnel arrives at the non-pickup stall, and the average value of the local congestion comprehensive influences of all non-pickup stalls in the candidate delivery path is calculated as the degree of way obstruction of the candidate delivery path; The candidate delivery path corresponding to the minimum degree of interference in the degrees of interference of all candidate delivery paths is taken as the final delivery path.

2. The Internet-based canteen delivery route planning method according to claim 1, characterized in that, The process of obtaining the passenger flow concentration situation includes: Obtaining the order quantity of each unit time period in the lunch peak period of the canteen stall every day, and obtaining the maximum order quantity therefrom; Obtaining the average value of the difference between the maximum order quantity and the order quantity in each unit time period on the same day; According to the average value, the passenger flow concentration situation of the canteen stall every day in the lunch peak period is obtained.

3. The Internet-based canteen delivery route planning method according to claim 1, characterized in that, The process of obtaining the busy period includes: Determining the order quantity of each unit time period in the lunch peak period of the canteen stall every day; Obtaining the average value of the order quantity in the same unit time period in the lunch peak period on all days to obtain the average order quantity of each unit time period; The target unit time period is formed into a busy period, and the target unit time period is a unit time period with an average order quantity greater than a preset reference order quantity.

4. The Internet-based canteen delivery route planning method according to claim 1, characterized in that, The process of obtaining the comprehensive pickup delay situation includes: Determining the pickup time length of the delivery personnel in the candidate busy period of the current stall; the current stall is any canteen stall, and the candidate busy period is any busy period of the current stall; Fusing the pickup time length and the local congestion comprehensive influence of the candidate busy period to obtain the comprehensive pickup delay situation of the candidate busy period.

5. The Internet-based canteen delivery route planning method according to claim 4, characterized in that, The process of obtaining the secondary order risk degree includes: obtaining the secondary order risk degree of the candidate busy period according to the pickup time length of the delivery personnel in the candidate busy period.

6. An Internet-based canteen delivery route planning system, characterized by comprising: Memory and processor; The memory is connected with the processor; The memory is used for storing program instructions; The processor is configured to implement the Internet-based canteen delivery path planning method according to any one of claims 1-5 when program instructions are executed.

Images