Pancake prefabricated paste intelligent distribution scheduling system based on real-time road condition

Abstract

The invention discloses a pancake prefabricated paste intelligent distribution scheduling system based on real-time road conditions, and relates to the technical field of food distribution scheduling, the system comprises a composite sensing module, a data processing module, a scheduling decision module and an execution module; a sensing system fusing multi-dimensional data is constructed through the composite sensing module, the problem that traditional single GPS positioning cannot comprehensively sense road conditions is solved, all-around control over traffic, weather and user behaviors is achieved, a road condition-time, weather and batter quality association and distribution delay probability prediction model is established through the data processing module, and the distribution delay probability prediction accuracy is improved. The problem of lack of data modeling analysis in the prior art is solved, a scientific basis is provided for scheduling, the problems of unreasonable resource allocation and weak risk handling capacity are solved by dynamically adjusting the distribution priority, allocating storage and triggering emergency order adding through the scheduling decision module, distribution delay and batter loss are finally reduced, and store operation is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of food delivery scheduling technology, specifically to an intelligent delivery scheduling system for pre-made pancake batter based on real-time traffic conditions. Background Technology

[0002] As a traditional specialty food, pancakes hold an important place in mass food consumption due to their crispy texture and diverse flavors. They are widely found in breakfast markets, snack stalls, and chain restaurants, satisfying the immediate eating needs of different consumers. Pre-made pancake batter is the core ingredient for making pancakes. It is made from flour, water, eggs, seasonings, and other basic components, pre-mixed and prepared, and can be directly used for pancake making. This effectively shortens the on-site operation time for pancake preparation, improves the efficiency of food service in restaurants, and is especially suitable for high-traffic commercial areas and the standardized operations of restaurant chains.

[0003] With the large-scale development of the pancake catering industry and the continuous increase in the number of chain stores, the demand for pre-made pancake batter delivery is also growing. The quality of pre-made pancake batter delivery directly affects the final taste and quality of the pancakes. If the delivery time is too long, the batter is prone to problems such as moisture loss, thickening, or spoilage, which in turn affects the spreading effect and eating experience of the pancakes. At the same time, timely and accurate delivery is also related to the normal operation of catering stores, avoiding store closures or loss of customers due to batter shortages. Therefore, building an efficient and intelligent pre-made pancake batter delivery and scheduling system is of great significance for ensuring batter quality, improving store operational efficiency, and enhancing customer satisfaction.

[0004] However, current delivery scheduling technology for pre-made pancake batter still has certain shortcomings. Traditional delivery scheduling relies heavily on GPS positioning for route planning, which can only obtain basic geographic location information and cannot fully perceive actual road conditions, such as traffic congestion and road construction. This leads to a disconnect between planned routes and actual delivery needs, easily resulting in delivery delays. It also fails to consider the impact of weather factors on delivery. Rain, high temperatures, and other weather conditions not only affect the speed and range of delivery vehicles but may also accelerate the deterioration of batter quality. Existing scheduling methods lack dynamic response mechanisms for weather changes, lack analysis and prediction of user behavior, and cannot identify high-incidence periods and high-risk areas for order cancellations. When the risk of delivery delays is high, it is difficult to take countermeasures in advance, which can easily lead to batter waste and order losses. Furthermore, the lack of a correlation model between road conditions and delivery time makes it impossible to adjust delivery priorities according to the road condition characteristics of different times and regions, resulting in unreasonable allocation of delivery resources and further exacerbating the delivery delay problem. Therefore, developing an intelligent delivery scheduling system for pre-made pancake batter based on real-time road conditions is of great significance. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions. It can build a perception system that integrates multi-dimensional data through a composite sensing module to achieve comprehensive control over traffic, weather, and user behavior. Through the data processing module, it establishes a correlation model between traffic conditions and time, weather, and batter quality, as well as a prediction model of delivery delay probability, providing a scientific basis for scheduling. Through the scheduling decision module, it dynamically adjusts delivery priorities, allocates warehousing, and triggers emergency orders to reduce delivery delays and batter loss, ensuring store operations.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a smart delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions, the system comprising: a composite sensing module, a data processing module, a scheduling decision module, and an execution module; The composite sensing module includes vehicle-mounted sensors, a municipal traffic API interface, a meteorological data interface, and an order data acquisition unit. It is used to acquire data such as road congestion index, average vehicle speed, accident location, rainfall, high temperature, and historical order generation time, delivery address, and cancellation time. The composite sensing module transmits the collected data to the data processing module. The data processing module is used to model the data transmitted by the composite sensing module, establish a road condition-time mapping model, a correlation model between meteorological data and batter quality, and a delivery delay probability prediction model. The data processing module transmits the model results to the scheduling decision module. The scheduling decision module is used to adjust delivery priorities, allocate storage points, and trigger emergency order mechanisms based on the model results transmitted by the data processing module. The scheduling decision module transmits the scheduling strategy to the execution module. The execution module includes a delivery task allocation unit, a vehicle scheduling unit, and a warehouse coordination unit, which are used to execute the scheduling strategy transmitted by the scheduling decision module.

[0007] Furthermore, in the composite sensing module, onboard sensors collect real-time location and speed data of delivery vehicles. The municipal traffic API interface retrieves road construction and temporary traffic control information for the corresponding area from the municipal traffic management platform at preset time intervals. The composite sensing module integrates the vehicle driving data with the municipal traffic data and transmits this data, along with rainfall and high-temperature data obtained from the meteorological data interface and historical order generation time, delivery address, and cancellation time data obtained from the order data collection unit, to the data processing module. The preset time interval is flexibly set according to the traffic information update frequency of different areas, covering different types of areas in the urban core and suburbs. Before transmitting the data, the composite sensing module calculates the traffic data fusion weight using a formula to quantify the proportion of vehicle driving data and municipal traffic data in the fusion process. The formula is: ,in, For traffic data fusion weighting, The real-time vehicle speed is collected by onboard sensors. The legal speed limit for the corresponding road section, The duration of the traffic event obtained from the municipal transportation API interface. The preset traffic data update cycle, and These are the weighting coefficients. Based on statistical data on the matching degree between vehicle speed and actual road conditions in the area over the past six months, The determination was based on statistical data on the impact of traffic incidents in the region on delivery timeliness over the past six months.

[0008] Furthermore, when the data processing module establishes the road condition-time mapping model, it first classifies the delivery area, delivery time, and actual time consumption data in the historical delivery records, and divides the data dimensions according to weekdays and non-weekdays, and morning, noon and evening peak periods. Then, it matches the real-time traffic flow data transmitted by the composite sensing module with the classified historical data, and generates road condition and delivery time correlation curves for different time periods and different areas through data fitting, forming a road condition-time mapping model. The time span of the historical delivery records covers at least the past three months to include road condition characteristics of different seasons.

[0009] Furthermore, the correlation model between meteorological data and batter quality in the data processing module incorporates batter quality detection parameters, including batter viscosity, moisture content, and pH value. The data processing module uses machine learning algorithms to analyze the relationship between delivery time and these parameters under different rainfall intensities and high-temperature durations, establishing a correspondence table between meteorological factors, delivery time, and batter quality as part of the model. The batter quality detection parameters are collected and recorded using professional testing equipment before warehousing and after delivery. During the analysis, the data processing module calculates the influence coefficient of meteorological factors on batter quality using a formula to assess the degree to which the delivery process affects batter quality under different meteorological conditions. The formula is: ,in, The influence coefficient of meteorological factors on the quality of dough. The rainfall intensity is obtained from the meteorological data interface. The difference between the real-time temperature and the optimal storage temperature for the batter is set based on the batter type. To predict delivery time, and The influence coefficient, Based on statistical data on the change rate of paste quality under different rainfall intensities over the past six months, The determination was based on statistical data on the change rate of paste quality under different temperature differences over the past six months.

[0010] Furthermore, when the data processing module constructs the delivery delay probability prediction model, it first extracts the delivery time period, delivery area, and traffic conditions corresponding to the cancelled orders in the historical orders to determine the correlation threshold between order cancellation and delivery delay. Then, it combines the real-time traffic data transmitted by the composite sensing module and calculates the delivery delay probability of the current time period and area through a probability statistical algorithm. When the delay probability exceeds the correlation threshold, the corresponding area is marked. The correlation threshold is determined based on the historical statistical data of order cancellation rates in different cities.

[0011] Furthermore, when adjusting delivery priorities, the scheduling decision module identifies periodically congested areas and time periods based on the road condition-time mapping model transmitted by the data processing module. This includes streets around schools during the morning rush hour and commercial areas during the evening rush hour. The scheduling decision module sorts the batter delivery tasks in this area according to delivery distance and store batter inventory, and allocates delivery tasks based on the sorting results. Store batter inventory is uploaded to the system backend in real time through store terminals. During the sorting process, the scheduling decision module calculates a delivery task priority score using a formula to differentiate the urgency of different delivery tasks and determine the execution order. The formula is: ,in, Score the priority of delivery tasks. This represents the current remaining batter inventory at the store. To maximize the store's batter inventory capacity This is the standard delivery distance for this area. This represents the actual delivery distance for the current delivery task. and This is the priority coefficient. This was determined based on statistics of the frequency of store closures due to insufficient inventory over the past six months. It is determined based on statistical data on delivery delay rates corresponding to different delivery distances over the past six months.

[0012] Furthermore, when the scheduling decision module allocates storage points, it combines the meteorological data transmitted by the data processing module with the prediction results of the correlation model of batter quality. If it predicts that rainstorms will shorten the range of electric vehicles, the scheduling decision module retrieves the batter inventory information of surrounding storage points and the distance data from each storage point to the delivery destination. Based on the information and data, it selects storage points to allocate delivery tasks. The scope of surrounding storage points is set to a preset radius area centered on the delivery destination.

[0013] Furthermore, when the scheduling decision module triggers the emergency order addition mechanism, it first obtains the number of orders to be delivered in the high-risk area, the location and load status of existing delivery vehicles, calculates the scheduling distance and required time for the backup vehicle, and then sends a scheduling instruction to the backup vehicle closest to the high-risk area. At the same time, it synchronizes the order information of the high-risk area to the navigation system of the backup vehicle, and the load status of the backup vehicle is fed back to the scheduling decision module in real time through the vehicle terminal.

[0014] Furthermore, after receiving the scheduling instruction transmitted by the scheduling decision module, the warehouse coordination unit of the execution module retrieves the batter specifications and quantity information of the corresponding delivery task, matches the batter storage location in the warehouse, generates a batter outbound list and sends it to the warehouse staff terminal, and updates the warehouse inventory data at the same time. The batter specifications include key information corresponding to the delivery task, such as the flavor type and packaging specifications of the batter.

[0015] Compared with existing technologies, this intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions has the following advantages: This invention constructs a perception system that integrates multi-dimensional data through a composite sensing module, solving the problem that traditional single GPS positioning cannot fully perceive road conditions. It enables comprehensive control over traffic, weather, and user behavior. Through the data processing module, it establishes a correlation model between road conditions and time, weather, and batter quality, as well as a prediction model for delivery delay probability, solving the problem of insufficient data modeling and analysis in existing technologies. This provides a scientific basis for scheduling. Through the scheduling decision module, it dynamically adjusts delivery priorities, allocates warehousing, and triggers emergency orders, solving the problems of unreasonable resource allocation and weak risk response capabilities. Ultimately, it reduces delivery delays and batter loss, ensuring store operations.

[0016] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 A schematic diagram of a smart delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions; Figure 2This is a flowchart of the intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] This invention provides an intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions. The system comprises four main modules: composite sensing, data processing, scheduling decision-making, and execution, forming a complete intelligent delivery and scheduling technical solution. (See also...) Figure 1 and Figure 2 The details are as follows: The composite sensing module consists of vehicle-mounted sensors, a municipal traffic API interface, a meteorological data interface, and an order data acquisition unit. The vehicle-mounted sensors collect the real-time location and speed of delivery vehicles; the municipal traffic API interface retrieves road construction and temporary traffic control information from the municipal traffic management platform at flexibly set preset time intervals; the meteorological data interface acquires rainfall and high-temperature data; and the order data acquisition unit acquires historical order creation time, delivery address, and cancellation time data. The module first merges vehicle movement and municipal traffic data, then transmits the merged data along with meteorological and order data to the data processing module. Before transmission, the traffic data fusion weight is calculated to quantify the fusion ratio of the two types of data.

[0021] The data processing module models the received data, constructing three types of models: First, a traffic-time mapping model, which first classifies historical delivery records by delivery area, time, and actual delivery time according to weekdays and non-weekdays, and morning, noon, and evening peak hours, and then matches real-time traffic flow data with the classified historical data to generate correlation curves, with historical records covering the past three months; Second, a meteorological data and batter quality correlation model, which introduces detection parameters such as batter viscosity, moisture content, and pH value to analyze the relationship between delivery time and quality parameters under rainfall intensity and high temperature duration, and establishes a corresponding relationship table, with parameters collected before warehouse departure and after delivery arrival; Third, a delivery delay probability prediction model, which extracts relevant data on canceled orders to determine correlation thresholds, combines real-time traffic conditions to calculate the current delay probability, and marks areas exceeding the threshold.

[0022] The scheduling decision module performs three operations based on the model results: adjusting delivery priorities, identifying periodic congestion areas and time periods, and allocating tasks according to delivery distance and store inventory balance, with store inventory uploaded in real time; allocating storage points, combining weather and batter quality model predictions, retrieving surrounding warehouse inventory and distance data to the destination to select storage points; and triggering an emergency order addition mechanism, obtaining the number of orders to be delivered in high-risk areas, vehicle location and load, calculating the dispatch distance and time for backup vehicles, sending instructions and synchronizing order information to the backup vehicle navigation system.

[0023] The execution module includes delivery task allocation, vehicle scheduling, and warehouse coordination units, and is responsible for executing the scheduling strategy. Upon receiving an instruction, the warehouse coordination unit retrieves the batter specifications and quantity for the delivery task, matches the storage location, generates an outbound manifest, sends it to the staff's terminal, and simultaneously updates the inventory data.

[0024] Example 1

[0025] This embodiment applies to a city-level pre-made batter delivery scenario for a chain pancake brand. This brand has three regional warehousing centers within the city, serving 20 stores located in the city's core commercial areas, around schools, and suburban residential areas. Peak delivery times are from 6:00 AM to 8:00 PM daily. The delivery needs to address the impact of morning rush hour traffic congestion around schools, midday order surges in commercial areas, evening traffic slowdowns in suburban areas, and sudden weather changes such as rain and high temperatures. Simultaneously, it must ensure the batter's consistent quality during delivery to avoid store stockouts or batter spoilage due to delivery delays. (See also...) Figure 1 and Figure 2 The intelligent delivery scheduling system based on the present invention can realize multi-dimensional data fusion perception, scientific modeling analysis and dynamic scheduling decision-making, and meet the needs of efficient delivery in this scenario.

[0026] During the system deployment phase, the installation and debugging of each component of the composite sensing module were completed first. Vehicle sensors were installed on all 20 delivery vehicles to collect vehicle location and speed data in real time. The municipal traffic management platform's open interface was used to configure the municipal traffic API interface, and preset time intervals were set according to the traffic information update frequency of different areas. Road construction and temporary traffic control information was obtained every 5 minutes in the urban core area and every 15 minutes in the suburbs. The system also connected to the meteorological department's data interface to obtain rainfall and high temperature data in real time. An order data collection unit was built in the system backend to synchronously store historical order generation time, delivery address, and cancellation time data for nearly one year.

[0027] During system initialization, the composite sensing module first calculates the traffic data fusion weights, determining the weight coefficients based on statistical data on the matching degree between vehicle speeds and actual road conditions in the urban core area and suburbs over the past six months. The weighting coefficients are determined based on statistical data on the impact of two types of regional traffic incidents on delivery timeliness over the past six months. Through formula The fusion weight is calculated to quantify the proportion of vehicle driving data and municipal traffic data in the fusion process. Then, the fused traffic data, along with meteorological data and historical order data, are transmitted to the data processing module.

[0028] After receiving the data, the data processing module begins to build three types of models. When building the traffic condition-time mapping model, it extracts historical delivery records from the past three months and divides the data into dimensions according to weekdays and non-weekdays, morning peak hours from 6-9 am, midday peak hours from 11 am to 1 pm, and evening peak hours from 5 pm to 8 pm. It then classifies the delivery area, delivery time, and actual delivery time data under different dimensions. Finally, it matches the real-time traffic flow data transmitted by the composite sensing module with the classified historical data and generates traffic condition and delivery time correlation curves for each time period and region through data fitting, forming a complete traffic condition-time mapping model.

[0029] When constructing the correlation model between meteorological data and batter quality, professional testing equipment was set up at the warehouse outlet and the receiving points of each store to collect the viscosity, moisture content, and pH value of the batter before and after leaving the warehouse as quality testing parameters. The data processing module used machine learning algorithms to analyze the relationship between delivery time and the changes in quality testing parameters under different rainfall intensities and high temperature durations. At the same time, the influence coefficient was determined based on the statistical data of the batter quality change rate under different rainfall intensities and temperature differences over the past six months. and Through formula Calculate the influence coefficient of meteorological factors on the quality of the batter, establish a table showing the correspondence between meteorological factors, delivery time and batter quality, and improve the model content.

[0030] When constructing a delivery delay probability prediction model, the delivery time, delivery area, and traffic conditions corresponding to canceled orders in historical orders are extracted. Based on the historical statistical data of order cancellation rate in the city over the past year, the correlation threshold between order cancellation and delivery delay is determined. Combined with real-time traffic data, the probability of delivery delay in the current time period and area is calculated using a probabilistic statistical algorithm. If the delay probability exceeds the correlation threshold, the area is marked.

[0031] After receiving the three types of model results output by the data processing module, the scheduling decision module performs dynamic scheduling operations. When adjusting delivery priorities, it identifies periodic congestion areas and time periods, such as streets around schools during morning rush hour and commercial areas during evening rush hour, based on the road condition-time mapping model. It also obtains the real-time batter inventory of each store through the store terminals and determines priority coefficients based on statistical data on the frequency of store closures due to insufficient inventory over the past six months and delivery delay rates corresponding to different delivery distances. and Through formula Calculate the priority score for each delivery task, sort them by score, delivery distance, and inventory level, and then allocate the delivery tasks. When allocating storage points, combine the prediction results of the correlation model between meteorological data and dough quality. If it is predicted that heavy rain will shorten the range of electric vehicles, retrieve the inventory information of surrounding storage points within a preset radius centered on each delivery destination, as well as the distance data from the storage points to the destination, and select the storage point with the closest distance and sufficient inventory to allocate the delivery task.

[0032] When the emergency order addition mechanism is triggered, for high-risk areas where the probability of delivery delay exceeds the threshold, the number of orders to be delivered in that area, the location and load status of existing delivery vehicles are obtained, and the dispatch distance and required time of the backup vehicles are calculated through the real-time feedback of the backup vehicle load data from the vehicle terminal. The dispatch instruction is sent to the nearest backup vehicle, and the order information of the high-risk area is synchronized to the navigation system of the backup vehicle.

[0033] After receiving the scheduling strategy from the scheduling decision module, the execution module assigns delivery tasks according to priority scores by the delivery task allocation unit. The vehicle scheduling unit dispatches the corresponding delivery vehicles and backup vehicles to the designated area. The warehousing coordination unit retrieves the specifications and quantity information of the batter for the corresponding delivery task, matches the batter storage location in the warehouse, generates a batter outbound list, and sends it to the warehouse staff's terminal. After the staff completes the outbound operation according to the list, the warehousing coordination unit updates the warehouse inventory data in real time to ensure the accuracy of the inventory data for subsequent scheduling decisions.

[0034] In summary, this embodiment, by applying a smart delivery scheduling system for pre-made pancake batter based on real-time traffic conditions in a city-level delivery scenario for a chain pancake brand, achieves comprehensive perception of traffic, weather, and user behavior. The construction of three types of models provides scientific data support for scheduling, dynamically adjusting delivery priorities, allocating storage points, and triggering emergency order mechanisms. This effectively solves the problems of unreasonable resource allocation and weak risk response capabilities in traditional delivery, reduces delivery delays, minimizes batter loss during delivery, ensures the normal operation of each store, and improves the timeliness and quality stability of batter delivery. It provides an efficient and feasible technical solution for the large-scale delivery of chain catering brands.

[0035] Example 2

[0036] This embodiment applies to the cross-district and county pre-made batter delivery scenario of a regional pancake chain brand. The brand has two core warehousing centers within the region, located in the eastern urban area and the western suburbs respectively, serving 15 stores. These stores are distributed around urban office building clusters, suburban rural markets, and transportation hubs at the borders of districts and counties. Daily peak order times are 7:00 AM to 9:00 AM in the office building clusters, 12:00 PM to 2:00 PM in the rural markets, and 6:00 PM to 8:00 PM at transportation hubs. The delivery needs to address issues such as frequent road construction between districts and counties, significant weather variations in rural areas, and heavy traffic and temporary traffic control at transportation hubs. Simultaneously, it must ensure that the batter quality is not affected by high temperatures and bumps during long-distance cross-district and county deliveries, and avoid supply disruptions to stores due to uneven distribution of warehousing resources or untimely vehicle dispatch. (See also...) Figure 1 and Figure 2 The intelligent delivery scheduling system based on the present invention can realize multi-scenario data collaborative perception, accurate modeling and analysis, and flexible scheduling decision-making, meeting the high-efficiency operation needs of this cross-county delivery scenario.

[0037] Before system startup, the component deployment and parameter configuration of the composite sensing module were completed. High-precision vehicle-mounted sensors were installed on 18 cross-district delivery vehicles to ensure real-time collection of vehicle location and speed on urban main roads, rural roads, and highways. Through the district-county linkage interface of the municipal traffic management platform, the municipal traffic API interface was configured, and preset time intervals were set according to the traffic information update frequency of different road sections: road construction and temporary traffic control information was obtained every 6 minutes on urban main roads, every 20 minutes on rural roads, and every 8 minutes on highways. The regional meteorological department's zoned data interface was connected to obtain rainfall and high temperature data in the eastern urban area, western suburbs, and district-county border areas in real time. An order data collection unit was built in the system backend to synchronously store nearly 18 months of cross-district historical order generation time, delivery address, and cancellation time data.

[0038] During system initialization, the composite sensing module first calculates the traffic data fusion weights, determining the weight coefficients based on statistical data on the matching degree between vehicle speeds and actual road conditions on various road sections over the past six months. The weighting coefficients are determined based on statistical data on the impact of traffic incidents on cross-district and county delivery timeliness over the past six months. Through formula The proportion of vehicle driving data and municipal traffic data fused together is quantified, and then the fused traffic data, along with regional meteorological data and cross-district historical order data, is transmitted to the data processing module.

[0039] After receiving the data, the data processing module constructs three core models in stages. When constructing the traffic condition-time mapping model, it extracts historical delivery records from the past three months across districts and counties, divides the data into dimensions according to weekdays and non-weekdays, and three peak order periods in the morning, noon and evening. It then classifies the delivery area, delivery time and actual consumption time data of each district and county under different dimensions. Finally, it accurately matches the real-time traffic flow data transmitted by the composite sensing module with the classified historical data, and generates the correlation curve between traffic conditions and delivery time for each time period and each district and county road segment through data fitting, forming a traffic condition-time mapping model covering cross-district and county scenarios.

[0040] When constructing the correlation model between meteorological data and batter quality, professional testing equipment was deployed at the outbound ports of the two core warehousing centers and the receiving points of 15 stores to collect the viscosity, moisture content, and pH value of the batter before and after its arrival as quality testing parameters. The data processing module used machine learning algorithms to analyze the relationship between delivery time and the changes in batter quality testing parameters under conditions of high temperatures in the eastern urban area, rainfall in the western suburbs, and diurnal temperature differences in the border areas between districts and counties. At the same time, the influence coefficient was determined based on statistical data of the rate of change in batter quality under different rainfall intensities and temperature differences over the past six months. and Through formula Calculate the influence coefficient of meteorological factors on the quality of dough in different regions, establish a table of correspondence between meteorological factors, cross-county delivery time and dough quality, and improve the model system.

[0041] When constructing a delivery delay probability prediction model, the delivery time, delivery area, and traffic conditions data corresponding to cross-district / county canceled orders in historical orders are extracted. Based on the historical statistical data of order cancellation rates in each district / county within the region over the past year, the correlation threshold between order cancellation and delivery delay is determined. Combined with real-time traffic data, the probability of delivery delay in the current time period and each district / county is calculated using a probabilistic statistical algorithm. If the delay probability exceeds the correlation threshold, the area is marked as high-risk.

[0042] After receiving the model results output by the data processing module, the scheduling decision module performs targeted scheduling operations. When adjusting delivery priorities, it identifies periodic congestion areas and time periods based on the road condition-time mapping model, such as road sections around office buildings in urban areas during the morning peak, road sections around rural markets during the midday peak, and road sections around transportation hubs during the evening peak. It obtains the real-time inventory of batter at each store through the store terminals and determines priority coefficients based on statistical data on the frequency of store closures due to insufficient inventory over the past six months and statistical data on delivery delay rates corresponding to different cross-district / county delivery distances. and Through formula Calculate the priority score for each delivery task, sort them by score in combination with delivery distance and inventory balance, and assign cross-district / county delivery tasks.

[0043] When allocating storage locations, the system combines meteorological data with the prediction results of a correlation model for batter quality. If it is predicted that rainfall in the western suburbs will extend delivery times or that high temperatures in the eastern urban area will accelerate changes in batter quality, the system retrieves inventory information and cross-county distance data of storage locations within a preset radius centered on each delivery destination. Priority is given to assigning delivery tasks to storage locations that are close by and have sufficient inventory, minimizing the impact of long-distance delivery on batter quality. When the emergency order addition mechanism is triggered, for high-risk areas where the probability of delivery delay exceeds a threshold, the system obtains the number of orders awaiting delivery in that area, the location and load status of existing delivery vehicles, and calculates the dispatch distance and required time for backup vehicles from nearby storage locations or vacant road sections to the high-risk area using real-time feedback from vehicle terminals. Dispatch instructions are then sent to the nearest backup vehicle, and order information from the high-risk area is synchronized to the backup vehicle's navigation system to ensure timely order delivery.

[0044] After receiving the scheduling strategy from the scheduling decision module, the execution module coordinates the execution of operations. The delivery task allocation unit assigns cross-district / county delivery tasks to corresponding vehicles according to priority scores; the vehicle scheduling unit dispatches regular delivery vehicles or backup vehicles to designated district / county road sections according to task requirements; the warehousing coordination unit retrieves the batter specifications, including flavor type, packaging specifications, and quantity information for the corresponding delivery task, matches them with the batter storage location in the warehouse, generates a batter outbound list, and sends it to the warehouse staff's terminal. After the staff completes the outbound operation according to the list, the warehousing coordination unit updates the inventory data of the two core warehousing centers in real time, providing accurate data support for subsequent cross-district / county scheduling decisions.

[0045] In summary, this embodiment, by applying a real-time traffic-based intelligent delivery scheduling system for pre-made pancake batter in a cross-county delivery scenario for a regional pancake chain brand, achieves comprehensive perception of traffic, weather conditions, and cross-regional user behavior in different districts and counties. The construction of three models provides accurate data for cross-county scheduling, dynamically adjusts delivery priorities, scientifically allocates storage points, and promptly triggers emergency order mechanisms. This effectively solves the problems of resource imbalance and weak ability to cope with complex road conditions and weather in traditional cross-county delivery, significantly reduces cross-county delivery delays, reduces batter quality loss during long-distance delivery, ensures the stable operation of stores in each district and county, and improves the efficiency and flexibility of cross-county delivery, providing reliable technical support for the large-scale expansion of regional catering brands.

[0046] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A smart delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions, characterized in that: The system includes: a composite sensing module, a data processing module, a scheduling decision module, and an execution module; The composite sensing module includes vehicle-mounted sensors, a municipal traffic API interface, a meteorological data interface, and an order data acquisition unit. It is used to acquire data such as road congestion index, average vehicle speed, accident location, rainfall, high temperature, and historical order generation time, delivery address, and cancellation time. The composite sensing module transmits the collected data to the data processing module. The data processing module is used to model the data transmitted by the composite sensing module, establish a road condition-time mapping model, a correlation model between meteorological data and batter quality, and a delivery delay probability prediction model. The data processing module transmits the model results to the scheduling decision module. The scheduling decision module is used to adjust delivery priorities, allocate storage points, and trigger emergency order mechanisms based on the model results transmitted by the data processing module. The scheduling decision module transmits the scheduling strategy to the execution module. The execution module includes a delivery task allocation unit, a vehicle scheduling unit, and a warehouse coordination unit, which are used to execute the scheduling strategy transmitted by the scheduling decision module.

2. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, In the composite sensing module, onboard sensors collect real-time location and speed data of delivery vehicles. The municipal traffic API interface retrieves road construction and temporary traffic control information for the corresponding area from the municipal traffic management platform at preset time intervals. The composite sensing module integrates the vehicle driving data with the municipal traffic data and transmits this data, along with rainfall and high-temperature data obtained from the meteorological data interface and historical order generation time, delivery address, and cancellation time data obtained from the order data collection unit, to the data processing module. The preset time interval is flexibly set according to the traffic information update frequency of different areas, covering different types of areas in both urban core areas and suburbs. Before transmitting the data, the composite sensing module calculates the traffic data fusion weight using a formula to quantify the proportion of vehicle driving data and municipal traffic data in the fusion process. The formula is: ,in, For traffic data fusion weighting, The real-time vehicle speed is collected by onboard sensors. The legal speed limit for the corresponding road section, The duration of the traffic event obtained from the municipal transportation API interface. The preset traffic data update cycle, and These are the weighting coefficients.

3. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, When establishing the traffic condition-time mapping model, the data processing module first classifies the delivery area, delivery time, and actual delivery time data in the historical delivery records, and divides the data dimensions according to weekdays and non-weekdays, and morning, noon and evening peak periods. Then, it matches the real-time traffic flow data transmitted by the composite sensing module with the classified historical data, and generates the correlation curve between traffic conditions and delivery time in different time periods and different areas through data fitting, thus forming a traffic condition-time mapping model. The time span of the historical delivery records covers at least the past three months.

4. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, The data processing module incorporates batter quality detection parameters into its meteorological data-battery quality correlation model. These parameters include batter viscosity, moisture content, and pH value. The module uses machine learning algorithms to analyze the relationship between delivery time and these parameters under different rainfall intensities and high-temperature durations, establishing a correspondence table between meteorological factors, delivery time, and batter quality as part of the model. The batter quality detection parameters are collected and recorded using specialized testing equipment before warehousing and after delivery. During the analysis, the module calculates the influence coefficient of meteorological factors on batter quality using a formula to assess the impact of the delivery process on batter quality under different meteorological conditions. The formula is: ,in, The influence coefficient of meteorological factors on the quality of dough. The rainfall intensity is obtained from the meteorological data interface. This is the difference between the real-time temperature and the suitable storage temperature for the batter. To predict delivery time, and This is the influence coefficient.

5. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, When the data processing module constructs the delivery delay probability prediction model, it first extracts the delivery time period, delivery area, and traffic conditions corresponding to the cancelled orders in the historical orders to determine the correlation threshold between order cancellation and delivery delay. Then, it combines the real-time traffic data transmitted by the composite sensing module and calculates the delivery delay probability of the current time period and area through a probability statistical algorithm. When the delay probability exceeds the correlation threshold, the corresponding area is marked. The correlation threshold is determined based on the historical statistical data of order cancellation rates in different cities.

6. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, When adjusting delivery priorities, the scheduling decision module identifies periodic congestion areas and time periods based on the road condition-time mapping model transmitted by the data processing module. This includes streets around schools during the morning rush hour and commercial areas during the evening rush hour. The scheduling decision module sorts the batter delivery tasks in this area according to delivery distance and store batter inventory, and allocates delivery tasks based on the sorting results. Store batter inventory is uploaded to the system backend in real time through store terminals. During the sorting process, the scheduling decision module calculates a delivery task priority score using a formula to differentiate the urgency of different delivery tasks and determine the execution order. The formula is: ,in, Score the priority of delivery tasks. This represents the current remaining batter inventory at the store. To maximize the store's batter inventory capacity This is the standard delivery distance for this area. This represents the actual delivery distance for the current delivery task. and This is the priority coefficient.

7. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, When the scheduling decision module allocates storage points, it combines the meteorological data transmitted by the data processing module with the prediction results of the correlation model of batter quality. If it predicts that rainstorms will shorten the range of electric vehicles, the scheduling decision module retrieves the batter inventory information of surrounding storage points and the distance data from each storage point to the delivery destination. Based on the information and data, it selects storage points to allocate delivery tasks. The scope of surrounding storage points is set to a preset radius area centered on the delivery destination.

8. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, When the scheduling decision module triggers the emergency order addition mechanism, it first obtains the number of orders to be delivered in the high-risk area, the location and load status of existing delivery vehicles, calculates the scheduling distance and required time for the backup vehicle, and then sends a scheduling instruction to the backup vehicle closest to the high-risk area. At the same time, the order information in the high-risk area is synchronized to the navigation system of the backup vehicle, and the load status of the backup vehicle is fed back to the scheduling decision module in real time through the vehicle terminal.

9. The intelligent delivery and scheduling system for pre-made pancake batter based on real-time traffic conditions according to claim 1, characterized in that, After receiving the scheduling instruction transmitted by the scheduling decision module, the warehouse coordination unit of the execution module retrieves the specifications and quantity information of the batter for the corresponding delivery task, matches the batter storage location in the warehouse, generates a batter outbound list and sends it to the warehouse staff terminal, and updates the warehouse inventory data at the same time.

Images