Intelligent bus combination scheduling method and device and storage medium

Abstract

The invention relates to an intelligent bus combined scheduling method and device and a storage medium, and relates to the field of scheduling optimization. Bus card swiping records are called to establish a card swiping station feature set, navigation station triggering is established to carry out navigation record matching analysis, and a space-time popularity prediction map is established; constructing a multi-source traffic flow data set to perform road congestion state prediction and construct a region-level road congestion state diagram; establishing a station passenger flow demand, route travel time and a congestion state according to the space-time popularity prediction map and the region-level road congestion state map; a multi-target scheduling channel is configured to take station passenger flow demands, route travel time and congestion states as input data, and route bus scheduling optimization is carried out through a passenger waiting time target, a scheduling cost target, a vehicle utilization rate target and a congestion influence target. The technical problems of insufficient data utilization, limited prediction precision and lack of flexibility of a scheduling strategy in bus scheduling are solved.

Description

Technical Field

[0001] The present invention relates to the field of scheduling optimization, and in particular to a smart bus combination scheduling method, device and storage medium. Background Art

[0002] With the acceleration of urbanization and the continuous growth of the population, public transportation plays a vital role in urban travel. Traditional bus scheduling methods often rely on manual experience and static data. Dispatchers manually set and adjust bus departure times and routes based on historical data and current road conditions. However, with the acceleration of urbanization and the increase in population mobility, traditional bus scheduling methods are unable to accurately reflect real-time passenger demand and road conditions, resulting in low bus operation efficiency and long passenger waiting times. Therefore, based on the development of smart cities, an intelligent bus scheduling method has been derived, which uses the internet to optimize bus scheduling.

[0003] However, existing bus dispatching methods also have certain limitations. On the one hand, existing bus dispatching methods lack real-time and accuracy. Due to limitations in data collection and processing methods, dispatchers struggle to obtain key information such as passenger demand and road congestion in a timely manner. As a result, dispatching plans often lag behind actual conditions and fail to respond promptly to changes in passenger demand and road conditions. On the other hand, existing bus dispatching methods lack intelligence and automation, and the dispatching process is easily influenced by human factors, resulting in suboptimal dispatching plans and an inability to maximize the utilization of bus resources and improve operational efficiency and service quality. Summary of the Invention

[0004] The present invention aims to solve the technical problems of insufficient data utilization, limited prediction accuracy and lack of flexibility in scheduling strategies in the existing bus scheduling technology, and provides a smart bus combination scheduling method, device and storage medium to solve them.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: In the first aspect, the present invention provides a smart bus combination scheduling method, which includes: calling bus card swiping records to establish a card swiping site feature set; establishing a navigation site trigger, using the navigation site trigger to perform navigation record matching analysis, and establishing a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set; constructing a multi-source traffic flow data set, using the multi-source traffic flow data set to predict road congestion status, and constructing a regional road congestion status map; establishing site passenger flow demand, line travel time and congestion status based on the spatiotemporal heat prediction map and the regional road congestion status map; configuring a multi-objective scheduling channel, using the site passenger flow demand, line travel time and congestion status as input data, and optimizing line bus scheduling through passenger waiting time targets, scheduling cost targets, vehicle utilization targets, and congestion impact targets to establish an optimization result; and performing bus scheduling management based on the optimization result.

[0006] In the second aspect, the present invention provides a smart bus combination scheduling device, which includes: an information call module for calling bus card swiping records and establishing a card swiping site feature set; a matching analysis module for establishing a navigation site trigger, using the navigation site trigger to perform navigation record matching analysis, and establishing a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set; a state prediction module for constructing a multi-source traffic flow data set, using the multi-source traffic flow data set to predict road congestion status, and constructing a regional road congestion status map; an information flow module for establishing site passenger flow demand, line travel time and congestion status based on the spatiotemporal heat prediction map and the regional road congestion status map; a channel configuration module for configuring a multi-objective scheduling channel, using the site passenger flow demand, line travel time and congestion status as input data, and optimizing line bus scheduling through passenger waiting time targets, scheduling cost targets, vehicle utilization targets, and congestion impact targets to establish an optimization result; a scheduling management module for performing bus scheduling management based on the optimization result.

[0007] In a third aspect, the present invention provides a computer device, comprising: a memory for storing executable instructions; The processor is used to implement the smart bus combination scheduling method described in the first aspect when running the executable instructions stored in the memory.

[0008] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, it implements a smart bus combination scheduling method as described in the first aspect.

[0009] The present invention has the following beneficial effects: by creating a feature set based on bus card swipe records, combining navigation record matching analysis to construct a spatiotemporal heat prediction map, and using multi-source traffic flow data to predict road congestion, a model for station passenger flow demand, route travel time, and congestion status is then established based on the prediction results. By configuring a multi-objective dispatch channel to optimize bus route dispatch, this method can effectively improve bus dispatch efficiency, reduce passenger wait times, lower dispatch costs, increase vehicle utilization, and mitigate the impact of traffic congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a smart public transportation combined dispatching method provided by the present invention.

[0011] Figure 2 This is a structural diagram of a smart bus combination dispatching device provided by the present invention.

[0012] Figure 3 A schematic structural diagram of a computer-readable storage medium provided by the present invention.

[0013] Explanation of the accompanying symbols: information calling module 11, matching analysis module 12, state prediction module 13, information flow module 14, channel configuration module 15, scheduling management module 16, computer-readable storage medium 600, first computer program 611. DETAILED DESCRIPTION

[0014] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0015] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0016] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0017] Example 1: like Figure 1 As shown, an embodiment of the present invention provides a smart bus combination scheduling method, the method comprising: S10: Call the bus card swiping record and establish a card swiping site feature set.

[0018] For example, when implementing smart bus combined scheduling, bus card swipe records are first accessed. This step forms the basis for data collection. These bus card swipe records contain information such as the boarding and alighting stations and times of each passenger's bus ride. By organizing and analyzing these swipe records, a swipe station feature set can be established. The swipe station feature set is a dataset containing multiple feature words that primarily reflects passengers' travel patterns and station preferences. For example, if a station sees a large number of passengers swiping their cards to board buses during peak hours in the morning and evening, this station might be labeled a "peak traffic hotspot." If a station consistently sees a fixed number of passengers swiping their cards to board and alight, this station might be labeled a "stable passenger flow station." During the process of establishing the swipe station feature set, adaptive adjustments can be made based on actual needs. For example, during specific holidays or events, passenger flow at certain stations may change significantly. In this case, the swipe station feature set can be updated to reflect these changes, thereby better adapting to actual conditions. Furthermore, through in-depth analysis of the swipe station feature set, potential travel patterns and trends can be identified, providing more accurate data support for subsequent bus scheduling.

[0019] S20: Establishing a navigation site trigger, using the navigation site trigger to perform navigation record matching analysis, and establishing a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set.

[0020] Optionally, a navigation site trigger is established, that is, when a passenger uses a specific software for navigation, the trigger mechanism is activated based on their own location and the selected destination. For example, a passenger in place A sets a destination to place B through a travel navigation software. At this time, based on their location information and destination information, the navigation site trigger mechanism is activated. Subsequently, the data obtained by this navigation site trigger is used to perform navigation record matching analysis. The system will compare the navigation records of different passengers and analyze the similarities and correlations therein, such as analyzing navigation records from similar departure points to similar destinations within the same time period. Furthermore, based on the results of the matching analysis, combined with the card swiping site feature set, it covers multi-dimensional information such as card swiping time, card swiping location, pedestrian flow around the site, and site type (such as whether it is a transfer station). For example, during the morning rush hour on weekdays, the card swiping records of a large transfer station will show characteristics such as concentrated card swiping time and large passenger flow. After integrating this information, a spatiotemporal heat prediction map is created. This map can intuitively show the popularity of public transportation stops in different areas and time periods. For example, it is predicted that the popularity of bus stops near a commercial district will increase significantly on weekend afternoons, providing a strong basis for public transportation operation scheduling, resource allocation and passenger travel planning.

[0021] S30: Constructing a multi-source traffic flow dataset, using the multi-source traffic flow dataset to predict road congestion status, and constructing a regional road congestion status map.

[0022] Specifically, a multi-source traffic flow dataset was constructed. This dataset integrates traffic data from various channels and types. For example, real-time traffic flow video data from traffic cameras can be analyzed to determine the number of vehicles passing through a specific road section per unit time. Vehicle speed data collected by induction coils installed on the road accurately reflects vehicle speeds. And GPS positioning data from floating vehicles (such as taxis and ride-hailing services) can track vehicle trajectories and dwell times. These data come from a wide range of sources and constitute the rich information dimensions of the multi-source traffic flow dataset.

[0023] After constructing a multi-source traffic flow dataset, it is used to predict road congestion. Data analysis algorithms and models are used to deeply mine and analyze multi-source data. For example, by analyzing the changing patterns of historical traffic flow data over different time periods, combined with current real-time traffic flow, vehicle speeds, and construction information on surrounding roads, road congestion can be predicted for the coming period. If a road has consistently high traffic volume during weekday morning and evening rush hours, and current real-time data shows a continued increase in traffic volume, and there is ongoing construction around the road, then congestion can be predicted for that road.

[0024] Finally, based on the results of the road congestion status prediction, a regional road congestion status map is constructed. The regional road congestion status map displays the congestion status of each road in a specific area in an intuitive graphical manner. For example, different colors are used to represent different degrees of congestion, red represents severe congestion, yellow represents mild congestion, and green represents smooth traffic. Taking the core business district of a city as an example, the regional road congestion status map can clearly show the congestion status of the main roads, secondary roads, and branches in the business district at different times. Based on this map, the traffic management department can reasonably adjust the duration of traffic lights, plan diversion routes, and guide vehicles to divert reasonably, thereby effectively alleviating regional traffic congestion problems.

[0025] S40: Establishing station passenger flow demand, route travel time and congestion status according to the spatiotemporal heat prediction map and the regional road congestion status map.

[0026] Further analysis is conducted based on the spatiotemporal heat prediction map and regional road congestion status map generated earlier. The spatiotemporal heat prediction map is generated based on a comprehensive analysis of various traffic-related data. It integrates information from both temporal and spatial dimensions, clearly displaying traffic activity in different areas at different time periods. For example, a commercial district may be more popular during the weekday evening rush hour, indicating high pedestrian and vehicle traffic. The regional road congestion status map, on the other hand, focuses on the road level, using real-time data and prediction models to visualize the congestion conditions of individual roads within a specific area. For example, a major arterial road is prone to congestion on rainy days, which is intuitively reflected in the map with different colors or symbols. Based on these two maps, passenger flow demand at each station can be further established. For example, using the spatiotemporal heat prediction map, if a bus stop is highly popular during a specific time period, it indicates a high potential passenger demand for the surrounding bus stops. For example, if the area housing a large shopping mall experiences a significant increase in popularity on weekends, it can be inferred that the passenger flow entering the nearby bus stops will also increase accordingly. At the same time, referring to the regional road congestion status map, if the road leading to the local bus is severely congested, it may cause some passengers who originally planned to take the bus to change their travel mode, which will also affect the passenger flow demand at the station. Combining these factors can more accurately establish a station passenger flow demand model.

[0027] Furthermore, for establishing route travel times, spatiotemporal heat prediction maps can provide trends in pedestrian and vehicle flows across different regions at different times, while regional road congestion maps can clearly identify the actual traffic conditions on each road. For example, a bus route passes through multiple areas. If one area experiences high traffic and congestion on surrounding roads during the morning rush hour, the bus's stop time in that area may be extended due to increased passenger boarding and alighting. Furthermore, bus speeds on ground routes or in areas connecting with other modes of transportation may also be reduced due to road congestion. Combining this information allows for an accurate travel time model for that route.

[0028] Regarding the establishment of congestion status, regional road congestion status maps themselves directly reflect road congestion. However, combining them with spatiotemporal heat prediction maps can further explore the causes and patterns behind congestion. For example, if a region experiences a sharp increase in popularity during a specific event, accompanied by congestion on surrounding roads, analysis can clearly indicate that the large influx of people and vehicles generated by the event is the cause of the congestion. This can then lead to the establishment of a more comprehensive congestion status assessment system, providing a strong basis for subsequent traffic diversion and management.

[0029] S50: Configure a multi-objective scheduling channel, take the passenger flow demand of the station, the route travel time and the congestion status as input data, optimize the route bus scheduling based on the passenger waiting time target, the scheduling cost target, the vehicle utilization rate target and the congestion impact target, and establish the optimization result.

[0030] S60: Performing public transportation dispatch management according to the optimization result.

[0031] Specifically, bus operations management requires the configuration of a multi-objective dispatch channel. This channel can be understood as an information processing hub that integrates key input data such as station passenger flow demand, route travel time, and congestion status. Station passenger flow demand reflects the passenger boarding and alighting needs of each bus station during different time periods. For example, during the weekday morning rush hour, passenger flow demand at stations near schools and office buildings will increase significantly. Route travel time covers the time it takes for a bus to travel from the starting station to the terminal, including stops at various stations and travel time on the route. Congestion status reflects the current traffic conditions of the road, such as when a major road is severely congested due to a traffic accident.

[0032] After these input data are imported into the multi-objective scheduling channel, bus scheduling optimization for each route is performed using passenger waiting time objectives, scheduling cost objectives, vehicle utilization objectives, and congestion impact objectives. The passenger waiting time objective aims to minimize passenger waiting time at stops and improve the travel experience, for example by more rationally scheduling bus departure intervals to reduce passenger waiting time. The scheduling cost objective focuses on reducing bus operating costs while ensuring service quality, such as rationally arranging the number of buses and routes to reduce unnecessary fuel consumption and vehicle wear. The vehicle utilization objective strives to ensure that every bus is fully utilized to avoid idle vehicles, such as flexibly allocating vehicles according to passenger demand to increase vehicle load factors. The congestion impact objective, as a reverse externality impact indicator, is used to assess the impact of bus additions on road congestion. For example, if a large number of buses rush into a congested section during peak hours, it may further exacerbate road congestion, so scheduling plans need to be optimized to reduce this negative impact.

[0033] For example, a bus route in a certain city passes through a commercial district, residential areas, and schools. During the morning rush hour, passenger demand is high at stops near schools. If conventional scheduling is used, the large number of buses entering this area could lead to increased congestion on surrounding roads. In this case, a multi-objective scheduling channel optimizes the bus schedule and routes, taking into account the four objectives above. For example, some buses may depart earlier or be redirected to less congested routes to balance passenger wait time, scheduling costs, vehicle utilization, and the impact on congestion.

[0034] Finally, the optimization results are generated, which include information such as optimized bus departure plans and routes. Based on this optimization result, bus dispatch management can be carried out to rationally arrange bus operations, improve the overall operating efficiency and service quality of the bus system, and provide passengers with a more convenient and comfortable travel environment.

[0035] In a preferred embodiment, the establishment of a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set includes: establishing site function labels, site co-occurrence probability labels, and site passenger density gradient labels; using the site function labels, site co-occurrence probability labels, and site passenger density gradient labels to predict the destination site of passengers boarding the site and establish a destination prediction result; using the destination prediction result to establish a passenger site path, and updating the destination prediction result and the passenger site path to the spatiotemporal heat prediction map.

[0036] In a specific embodiment, the process of constructing a spatiotemporal heat prediction map requires the establishment of a series of labels. Station function labels are used to identify the primary attributes of a station, such as categorizing stations into office areas, residential areas, commercial districts, and schools. These areas are typically locations with a high probability of passengers boarding and alighting. For example, stations located in areas with a high concentration of office buildings are labeled "office areas." During peak hours in the morning and evening, these stations experience high passenger boarding and alighting volumes. Station co-occurrence probability labels reflect passenger travel habits. For example, a passenger often boards the bus at Station A and then stops near Station B. By collecting and analyzing large amounts of passenger travel data, co-occurrence relationships between these stations can be discovered. For example, many office workers may board the bus at Station A near their residence and disembark at Station B near their workplace every day. This regular travel behavior is captured by the station co-occurrence probability labels. Station passenger density gradient labels are defined based on the difference in the number of passengers boarding and alighting at a station within a specific time period. If the number of passengers disembarking at a station significantly exceeds the number of passengers boarding during a specific period, then the station is likely a disembarkation point. For example, on weekend afternoons, a large number of passengers may get off the bus at a station near a large shopping mall to go shopping, while relatively few people will get on the bus. At this time, the station will show obvious characteristics of a get-off point.

[0037] After establishing these labels, the station function label, station co-occurrence probability label, and station passenger density gradient label are used to predict the destination station for passengers who board at a station. For example, for a passenger who boards at an office district station, the station function label indicates that this area is mostly populated by office workers. Furthermore, the station co-occurrence probability label indicates that many office workers with similar travel habits frequently stop near a certain commercial district station. Furthermore, the station passenger density gradient label indicates that this commercial district station has a high number of alightings during the corresponding time period. Combining this information, it can be predicted that this passenger's destination station is likely to be that commercial district station. In this way, destination station predictions can be performed for a large number of boarding passengers, ultimately establishing a destination prediction result.

[0038] Next, the destination prediction results are used to establish the passenger station path. For example, if a passenger is known to get on the bus at the office area station and is predicted to have a destination station at the business district station, then it can be determined that the passenger's station path is from the office area station to the business district station. Finally, the destination prediction results and the passenger station path are updated to the spatiotemporal heat prediction map. As new destination prediction results and passenger station paths are continuously added, the spatiotemporal heat prediction map can more accurately reflect the travel heat and flow direction of passengers in different areas and time periods. For example, through the updated spatiotemporal heat prediction map, it can be clearly seen that during the morning rush hour on weekdays, the path heat from the residential area station to the office area station is higher, while on weekends, the path heat from the residential area station to the business district station will increase significantly, which provides strong data support for traffic planning, resource allocation, etc.

[0039] In a preferred embodiment, the route bus scheduling optimization based on passenger waiting time objectives, scheduling cost objectives, vehicle utilization objectives, and congestion impact objectives includes: identifying shared stop buses for the current route buses and establishing a collaborative scheduling bus set; calling the passenger stop path, using the passenger stop path to perform shared scheduling matching of the collaborative scheduling bus set, generating a shared scheduling matching result, and configuring a shared starting station; establishing an extended objective function, the extended objective function being a shared collaborative evaluation function, the evaluation characteristics of which include shared waiting time and shared congestion comfort; obtaining the predicted time node and predicted in-vehicle passenger status of the collaborative scheduling bus set to the shared starting station; performing combined scheduling impact compensation of the extended objective function based on the predicted time node and the predicted in-vehicle passenger status, and completing the route bus scheduling optimization using the combined scheduling impact compensation.

[0040] Optionally, when optimizing bus dispatch for a route, it's first necessary to identify buses that share common stops on the current route. This involves analyzing buses on different routes to identify those that stop at certain common stops, and then establishing a set of buses for collaborative dispatch. For example, in a city bus network, if there are two routes, one serving route 1 and the other serving route 3, and both stop at stops A and B, the buses on these two routes can be included in the set of buses for collaborative dispatch.

[0041] Next, the passenger's stop paths are retrieved and used to match the shared dispatch of the collaboratively dispatched bus set. For example, if a passenger needs to travel from stop C to stop D, and there are buses in the collaboratively dispatched bus set that can pass through stops C and D, the system will match them, generate a shared dispatch match, and configure a shared start stop. For example, if it finds that both bus routes 1 and 3 can meet the travel demand from C to D and can use stop A as a shared start stop, then the corresponding match will be generated.

[0042] Next, an extended objective function is established: a shared collaborative evaluation function whose evaluation features include shared waiting time and shared crowding comfort. Shared waiting time refers to the time passengers wait for the bus at the shared starting stop, while shared crowding comfort reflects the impact of bus crowding on the passenger experience. For example, if the shared waiting time is too long or the bus is too crowded, the value of the shared collaborative evaluation function will be affected.

[0043] Next, the system obtains the predicted time nodes and predicted passenger status of the coordinated bus set arriving at the shared starting station. The predicted time node estimates the arrival time of the bus based on factors such as bus speed and road conditions. The predicted passenger status uses historical data and real-time information to predict the number and distribution of passengers at the time of arrival. For example, an intelligent algorithm predicts that bus No. 1 will arrive at Station A in 10 minutes and will be crowded with a large number of passengers.

[0044] Finally, the combined scheduling impact compensation of the extended objective function is performed based on the predicted time node and the predicted passenger status. This compensation mechanism can be understood as a reward or penalty term in multi-objective optimization, or it can be used independently as a local fine-tuning. For example, if a bus is predicted to be very crowded when it arrives at the shared starting station, then during the scheduling optimization process, a certain penalty will be imposed on the option to select that bus, reducing its probability of being selected. Conversely, if the bus is predicted to be relatively spacious and the waiting time will be short when it arrives, then a certain reward will be given, increasing its probability of being selected. In this way, the above data is continuously updated, and the route bus scheduling optimization is ultimately completed, achieving more reasonable and efficient bus scheduling management and improving passengers' travel experience.

[0045] In a preferred embodiment, the route bus scheduling is optimized according to the passenger waiting time target, scheduling cost target, vehicle utilization target, and congestion impact target, and the optimization result is established, including: predicting the passenger arrival time distribution at the station according to the passenger flow demand at the station, and establishing the passenger aggregation curve at the station; predicting the bus arrival time based on the scheduling data in the solution pool, performing overlapping calculation based on the arrival time prediction result and the passenger aggregation curve at the station, and establishing the average waiting time; predicting the fuel consumption, driving time, and off-peak empty driving scheduling cost of the scheduling data in the solution pool according to the route travel time and congestion status, and generating a scheduling cost prediction result; and optimizing the bus scheduling under the passenger waiting time target and the scheduling cost target according to the average waiting time and scheduling cost prediction results.

[0046] For example, during the process of optimizing bus route scheduling and establishing the optimal results, it is necessary to further predict the arrival time distribution of passengers at each station based on station passenger flow demand, thereby constructing a station passenger concentration curve. Station passenger flow demand reflects the travel demand of passengers at stations during different time periods. For example, during the weekday morning rush hour, passenger flow demand at stations near schools and office buildings increases significantly. By analyzing multiple sources of information, such as historical passenger flow data and real-time location data, the number of passengers arriving at stations at different times can be predicted, thereby creating a station passenger concentration curve. For example, between 7:00 and 8:00 a.m., the number of passenger arrivals near a subway station increases rapidly, peaks at 8:00 a.m., and then gradually decreases. This curve intuitively illustrates the passenger concentration at that station during this time. Next, bus arrival times are predicted based on the scheduling data in the solution pool. The solution pool contains data for various possible bus scheduling scenarios. Using prediction models and algorithms, combined with factors such as road conditions and traffic flow, the arrival times of buses at various stations can be predicted under different scheduling scenarios. For example, based on historical travel data for a particular route and current real-time road information, it is predicted that a particular bus will arrive at a station at 9:00 a.m. under a specific scheduling scenario. Then, an average wait time is calculated by overlaying the arrival time predictions with the passenger cluster curve at the station. This means comparing the bus arrival time predictions with the passenger cluster curve to calculate the average wait time at the station. For example, if a bus is predicted to arrive at a station at 9:00, and the passenger cluster curve shows a large number of passengers arriving between 8:50 and 9:00, the average wait time for these passengers can be calculated.

[0047] At the same time, the dispatching data in the solution pool is used to predict fuel consumption, driving time, and off-peak idle driving costs based on route travel time and congestion status, generating a dispatching cost forecast. Route travel time is affected by factors such as road congestion and station stop times, while congestion status directly reflects road traffic conditions. By analyzing these factors, fuel consumption, driving time, and off-peak idle driving costs can be predicted for different dispatching scenarios. For example, when traveling on congested roads, a bus's fuel consumption increases and driving time is prolonged. During off-peak hours, some buses may run idle, all of which increase dispatching costs.

[0048] Finally, based on the average wait time and dispatch cost predictions, bus dispatch optimization is performed under passenger wait time and dispatch cost objectives. With the goals of minimizing average wait time and minimizing dispatch cost, dispatch solutions in the solution pool are evaluated and screened. For example, while one dispatch solution may shorten average passenger wait times, it may have excessively high dispatch costs; while another solution may have lower dispatch costs but result in longer average passenger wait times. By comprehensively considering these two objectives, an optimization algorithm is used to continuously adjust and optimize the dispatch solutions, ultimately finding the optimal bus dispatch solution that meets both passenger wait time and dispatch cost objectives. This optimization result is then established, providing a more scientific and rational basis for dispatch decision-making in bus operations.

[0049] In a preferred embodiment, the bus scheduling optimization under the passenger waiting time target and the dispatching cost target based on the average waiting time and the dispatching cost prediction result includes: obtaining the maximum capacity of the bus; predicting the vehicle passenger density of the dispatching data in the solution pool to establish a passenger density prediction result; calculating the effective passenger rate based on the maximum capacity and passenger density prediction results to generate an effective passenger rate calculation result; obtaining the arrival time of the congested section based on the dispatching data in the solution pool, predicting the congestion state using the arrival time, and establishing a basic congestion state; fitting the intervention increment under the basic congestion state based on the bus length characteristics and the stop characteristics to generate a fitted congestion state; and using the average waiting time, the dispatching cost prediction result, the effective passenger rate calculation result, and the fitted congestion state to perform bus scheduling optimization under the passenger waiting time target, the dispatching cost target, the vehicle utilization target, and the congestion impact target.

[0050] Preferably, during the bus scheduling optimization process under passenger wait time and scheduling cost objectives, the maximum bus capacity is determined. For example, a certain bus model can carry a maximum of 80 passengers. Next, a vehicle passenger density prediction is performed on the scheduling data in the solution pool. The solution pool contains data on various possible bus scheduling scenarios. By analyzing historical passenger flow data, boarding and alighting patterns at stations, and other information, the passenger density at various stations and routes under different scheduling scenarios is predicted, thereby generating a passenger density prediction. For example, it is predicted that during peak hours on a certain route, the passenger density on a particular bus will increase significantly after passing a station near a school. The effective occupancy rate is calculated based on the bus's maximum capacity and passenger density predictions. The effective occupancy rate reflects the ratio of the number of passengers actually carried by the bus to its maximum capacity, generating a calculated effective occupancy rate. For example, if a bus has a maximum capacity of 80 passengers and actually carries 60 passengers during a certain period, the effective occupancy rate is 75%.

[0051] At the same time, the arrival times of congested sections are obtained based on the dispatch data in the solution pool. These arrival times are used to predict the congestion status and establish a basic congestion status. Specifically, by analyzing the arrival time of the bus and the historical congestion data of the congested section, the congestion level of the section at the time of the bus's arrival is predicted, including indicators such as traffic speed and queue length. For example, it is predicted that when a bus arrives at a congested section at 5 pm, the traffic speed on that section will drop to 10 kilometers per hour.

[0052] Furthermore, we perform incremental intervention fitting under the baseline congestion state based on bus length and stop characteristics. Bus length affects the space it occupies on the road, while stop characteristics such as stop time and location affect surrounding traffic flow. By considering these factors, we measure the degree of negative perturbation the addition of a bus to a particular road section's average per-unit-time capacity (velocity, density, load, etc.) to generate a fitted congestion state. For example, a long bus with prolonged stop times joining a congested section will further reduce traffic speed and capacity.

[0053] Finally, the system uses average waiting time, dispatch cost predictions, effective ridership calculations, and congestion profiles to optimize bus dispatch based on passenger waiting time, dispatch cost, vehicle utilization, and congestion impact objectives. With the goals of minimizing average waiting time, minimizing dispatch cost, maximizing vehicle utilization, and minimizing congestion impact as the primary objectives, the system comprehensively evaluates and screens dispatch solutions from the solution pool. For example, while one dispatch solution may shorten average passenger waiting time, it may also incur excessive dispatch costs and significantly impact congested sections. Meanwhile, another solution strikes a good balance across all these objectives. By continuously optimizing and adjusting dispatch solutions, the system ultimately finds the optimal bus dispatch solution that meets these multiple objectives, achieving efficient, orderly, and sustainable bus operations.

[0054] In a preferred embodiment, the bus dispatch management based on the optimization result includes: continuously adding matching records triggered by navigation sites to establish site incremental data; using the site incremental data to fine-tune the optimization result scheduling to complete the bus dispatch management.

[0055] Furthermore, after completing the route bus dispatch optimization search and obtaining the optimization results, the bus dispatch management stage begins. First, new matching records triggered by navigation stations will be continuously added. This process is achieved by using a mechanism triggered by passengers using specific navigation software based on their own positioning and destination selection. For example, when a large number of passengers use navigation software to go to or leave a popular business district during peak hours in the morning and evening, the system will accurately capture the association information between different stations and the business district based on these navigation records, such as which stations have the most passengers going to the business district, and which stations are mainly used by passengers from the business district. These newly generated matching records constitute the incremental station data. This incremental station data is real-time and dynamic, reflecting the latest changes in passenger travel demand. For example, during a large-scale event, the navigation trigger records of stations around the event will increase sharply. By continuously collecting the incremental station data formed by these records, the impact of the event on the surrounding public transportation travel demand can be captured in a timely manner.

[0056] Next, we use the incremental data from stations to fine-tune the scheduling of the optimization results. Since the optimization results are based on a certain time and data, and the actual public transportation travel demand is constantly changing, it is necessary to adjust the optimization results in a timely manner based on the incremental data from stations. For example, if the incremental data from stations shows that the number of passengers boarding at a certain station has increased significantly recently, and the destinations of these passengers are relatively concentrated, then we can fine-tune the bus departure intervals and vehicle allocation for the station in the optimization results, increase the number of buses heading to the destination, and better meet the travel needs of passengers.

[0057] In this way, the optimization results are continuously fine-tuned according to the real-time changing incremental data of stations, which can ensure that bus scheduling management always matches actual travel needs, improve the efficiency of bus operations and service quality, complete scientific and reasonable bus scheduling management, and provide passengers with a more convenient and comfortable bus travel experience.

[0058] In a preferred embodiment, the establishment of the spatiotemporal heat prediction map includes: identifying passenger entities and establishing specific passenger groups; setting scheduling preferences for the specific passenger groups, and adding the scheduling preferences to the spatiotemporal heat prediction map.

[0059] Specifically, in the process of constructing the spatiotemporal heat prediction map, it is also necessary to identify the main passenger groups. The purpose of identifying the main passenger groups is to distinguish different groups from the large number of passengers, such as office workers, students, senior citizens, and tourists. By analyzing multi-dimensional data such as passengers' travel time, travel location, and consumption habits, data mining and machine learning algorithms can accurately identify specific passenger groups. For example, by analyzing the frequent boarding of passengers at residential stations during the weekday morning rush hour and disembarking at stations near commercial areas or office buildings, these passengers can be identified as office workers.

[0060] Next, set the dispatch preferences for specific passenger groups. Different passenger groups have different preferences for bus dispatch due to differences in travel purposes, time sensitivity, and other factors. Commuters are generally more concerned with the punctuality and comfort of buses, hoping to reach their destinations quickly and conveniently during rush hour. Therefore, they may prefer buses with short departure intervals and a better interior environment. Students, on the other hand, may be more sensitive to fares and also hope that buses cover all areas surrounding their schools. Seniors may place greater emphasis on bus safety and ease of boarding and alighting. Based on these characteristics, set corresponding dispatch preferences for specific passenger groups.

[0061] Finally, dispatch preferences are added to the spatiotemporal heat forecast map. The spatiotemporal heat forecast map itself is a comprehensive tool that reflects traffic heat in different areas during different time periods. By incorporating the dispatch preferences of specific passenger groups, the forecast map can more accurately reflect the travel needs and preferences of different passenger groups. For example, on the spatiotemporal heat forecast map, areas and time periods with a high concentration of office workers can be marked with heat information that places high demands on bus punctuality and comfort; areas with a high concentration of students can highlight relevant heat information such as fares and coverage. In this way, bus operation management departments can refer to the spatiotemporal heat forecast map when formulating dispatch plans and rationally arrange bus routes, departure intervals, and bus models based on the dispatch preferences of different passenger groups, thereby improving the quality and efficiency of bus services and better meeting passengers' travel needs.

[0062] The embodiment of the present invention provides a smart bus combined dispatching method, which has at least the following technical effects: 1. By incorporating bus card swipe records into a feature set of swipe station locations, this method combines matching analysis triggered by navigation stations to create a spatiotemporal heat map. Furthermore, a multi-source traffic flow dataset is constructed to predict road congestion and build regional road congestion maps. The integration of multi-source data enables more accurate predictions of station passenger flow demand, route travel time, and congestion status, providing a reliable data foundation for subsequent scheduling optimization and helping to improve the scientific and rational nature of bus scheduling. For example, bus card swipe records can be used to understand passenger boarding and alighting patterns, navigation records can reflect real-time travel demand, and traffic flow data can accurately grasp road congestion conditions. Combining this information provides a more comprehensive understanding of the bus operating environment.

[0063] 2. Configure a multi-objective scheduling channel to optimize bus route scheduling based on passenger waiting time, scheduling costs, vehicle utilization, and congestion impact. During the optimization process, not only the current bus routes are considered, but also shared stop buses are identified, a collaborative scheduling bus set is established, shared scheduling is matched based on passenger stop paths, and an extended objective function is used to compensate for the impact of combined scheduling. This multi-objective optimization and collaborative scheduling approach can fully consider multiple aspects of bus operations, achieve efficient resource utilization and maximize overall benefits, and improve the quality and efficiency of bus services. For example, collaborative scheduling can reduce idle vehicle driving, reduce scheduling costs, and shorten passenger waiting times.

[0064] 3. When managing bus dispatch based on the optimization results, matching records triggered by navigation stations are continuously added to establish incremental station data, which is then used to fine-tune dispatching of the optimization results, enabling dynamic adjustments to bus dispatch to adapt to changing travel needs. Furthermore, when establishing a spatiotemporal heat map, passenger entities are identified, specific passenger groups are established, and their dispatch preferences are set. These preferences are then added to the prediction map to provide personalized bus services for different passenger groups. This dynamic adjustment and personalized service can better meet the diverse needs of passengers, improve passenger satisfaction, and enhance the attractiveness and competitiveness of public transportation. For example, bus dispatch plans tailored to the travel characteristics of different groups, such as office workers and students, can be provided.

[0065] Example 2: like Figure 2 As shown, based on the same inventive concept as the smart bus combination dispatching method provided in Example 1, an embodiment of the present invention further provides a smart bus combination dispatching device, the device comprising: The information calling module 11 is used to call the bus card swiping records and establish a card swiping site feature set.

[0066] The matching analysis module 12 is used to establish a navigation site trigger, use the navigation site trigger to perform navigation record matching analysis, and establish a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set.

[0067] The state prediction module 13 is used to construct a multi-source traffic flow dataset, use the multi-source traffic flow dataset to predict the road congestion state, and construct a regional road congestion state map.

[0068] The information flow module 14 is used to establish the site passenger flow demand, route travel time and congestion status according to the spatiotemporal heat prediction map and the regional road congestion status map.

[0069] The channel configuration module 15 is used to configure a multi-objective scheduling channel, taking the passenger flow demand of the station, the route travel time and the congestion status as input data, and optimizing the route bus scheduling based on the passenger waiting time target, the scheduling cost target, the vehicle utilization target and the congestion impact target to establish the optimization result.

[0070] The dispatch management module 16 is used to perform public transportation dispatch management according to the optimization result.

[0071] Furthermore, the matching analysis module 12 is further configured to perform the following steps: Establish site function labels, site co-occurrence probability labels, and site passenger density gradient labels; use the site function labels, site co-occurrence probability labels, and site passenger density gradient labels to predict the destination site of passengers boarding the site, and establish a destination prediction result; use the destination prediction result to establish a passenger site path, and update the destination prediction result and the passenger site path to the spatiotemporal heat prediction map.

[0072] Furthermore, the channel configuration module 15 is further configured to perform the following steps: Identify the shared stop buses for the current bus route and establish a collaborative dispatch bus set; call the passenger stop path, use the passenger stop path to perform shared dispatch matching for the collaborative dispatch bus set, generate a shared dispatch matching result, and configure a shared starting stop; establish an extended objective function, which is a shared collaborative evaluation function, and the evaluation characteristics include shared waiting time and shared congestion comfort; obtain the predicted time node and predicted passenger status of the collaborative dispatch bus set to the shared starting stop; perform combined dispatch impact compensation of the extended objective function according to the predicted time node and the predicted passenger status, and use the combined dispatch impact compensation to complete the route bus dispatch optimization.

[0073] Furthermore, the channel configuration module 15 is further configured to perform the following steps: The passenger arrival time distribution at the station is predicted based on the passenger flow demand at the station, and a passenger aggregation curve for the station is established; the bus arrival time is predicted based on the scheduling data in the solution pool, and an overlapping calculation is performed based on the arrival time prediction result and the passenger aggregation curve at the station to establish an average waiting time; the fuel consumption, driving time, and off-peak idle driving scheduling cost of the scheduling data in the solution pool are predicted based on the line travel time and congestion status, and a scheduling cost prediction result is generated; and bus scheduling is optimized under the passenger waiting time target and scheduling cost target based on the average waiting time and scheduling cost prediction results.

[0074] Furthermore, the channel configuration module 15 is further configured to perform the following steps: Obtain the maximum capacity of the bus; predict the vehicle passenger density of the dispatching data in the solution pool and establish a passenger density prediction result; calculate the effective passenger rate based on the maximum capacity and passenger density prediction results to generate an effective passenger rate calculation result; obtain the arrival time of the congested section based on the dispatching data in the solution pool, use the arrival time to predict the congestion state and establish a basic congestion state; perform intervention increment fitting under the basic congestion state based on the length characteristics and stop characteristics of the bus to generate a fitted congestion state; use the average waiting time, dispatching cost prediction result, effective passenger rate calculation result and fitted congestion state to optimize bus dispatch under the passenger waiting time target, dispatching cost target, vehicle utilization target and congestion impact target.

[0075] Furthermore, the scheduling management module 16 is further configured to perform the following steps: Continuously add matching records triggered by navigation sites to establish site incremental data; use the site incremental data to fine-tune the optimization results and complete bus scheduling management.

[0076] Furthermore, the matching analysis module 12 is further configured to perform the following steps: Identify the passenger body and establish a specific passenger group; set the scheduling preference for the specific passenger group, and add the scheduling preference to the spatiotemporal heat prediction map.

[0077] Through the above detailed description of a smart bus combination dispatching method in this specification, those skilled in the art can clearly understand a smart bus combination dispatching device in this embodiment. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0078] Example 3: An embodiment of the present invention provides a computer device, comprising: a memory for storing executable instructions; The processor is used to implement a smart bus combination scheduling method as described in Example 1 when running the executable instructions stored in the memory.

[0079] It should be noted that the processor may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor may be implemented in hardware using at least one of the following: digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory storage devices. In some embodiments, the computer device may optionally include an input interface and an output interface. The processor, memory, and the input interface and output interface may be connected via a bus or signal lines. Each peripheral device may be connected to the input interface and output interface via a bus, signal lines, or circuit board. The input interface and output interface may be used to connect at least one input / output-related peripheral device to the processor and memory. In some embodiments, the processor, memory, input interface, and output interface are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor, memory, input interface, and output interface can be implemented on a separate chip or circuit board, which is not limited in this embodiment of the present invention.

[0080] Example 4: See also Figure 3 , Figure 3 Schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 3 As shown, this embodiment provides a computer-readable storage medium 600 on which a second computer program 611 is stored. When the second computer program 611 is executed by a processor, it implements a smart bus combination scheduling method as in Example 1.

[0081] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0082] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatuses, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0083] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0084] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0086] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A smart bus combined dispatching method, characterized in that: The method comprises: Call the bus card swiping records and establish the card swiping site feature set; Establishing a navigation site trigger, using the navigation site trigger to perform navigation record matching analysis, and establishing a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set; Constructing a multi-source traffic flow dataset, using the multi-source traffic flow dataset to predict road congestion status, and constructing a regional road congestion status map; Establishing station passenger flow demand, route travel time and congestion status based on the spatiotemporal heat prediction map and the regional road congestion status map; A multi-objective scheduling channel is configured, and the passenger flow demand of the station, the travel time of the route and the congestion status are used as input data. The route bus scheduling is optimized based on the passenger waiting time target, the scheduling cost target, the vehicle utilization rate target and the congestion impact target, and an optimization result is established; Public transportation dispatch management is performed according to the optimization result.

2. A smart bus combination dispatching method according to claim 1, characterized in that: The process of establishing a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set includes: Establish station function labels, station co-occurrence probability labels, and station passenger density gradient labels; Using the station function labels, station co-occurrence probability labels, and station passenger density gradient labels, the destination station of passengers boarding the station is predicted, and a destination prediction result is established; The destination prediction result is used to establish a passenger station path, and the destination prediction result and the passenger station path are updated to the spatiotemporal heat prediction map.

3. A smart bus combination dispatching method as claimed in claim 2, characterized in that: The optimization of bus route scheduling based on passenger waiting time target, scheduling cost target, vehicle utilization target, and congestion impact target includes: Identify buses sharing common stops on the current bus route and establish a coordinated dispatch bus set; Calling the passenger stop path, using the passenger stop path to perform shared scheduling matching of the collaboratively scheduled bus set, generating a shared scheduling matching result, and configuring a shared starting stop; Establishing an extended objective function, wherein the extended objective function is a shared collaborative evaluation function, and the evaluation characteristics include shared waiting time and shared congestion comfort; Obtain the predicted time node and predicted passenger status of the coordinated dispatched bus set to the shared starting station; The combined scheduling impact compensation of the extended objective function is performed according to the predicted time node and the predicted passenger status, and the route bus scheduling optimization is completed using the combined scheduling impact compensation.

4. The intelligent public transportation combined dispatching method according to claim 1, characterized in that: The optimization of bus route scheduling is performed based on the passenger waiting time target, scheduling cost target, vehicle utilization target, and congestion impact target, and the optimization result is established, including: Predicting the passenger arrival time distribution at the station based on the passenger flow demand at the station and establishing a passenger aggregation curve at the station; Based on the dispatch data in the solution pool, the bus arrival time is predicted. The average waiting time is established by performing overlapping calculation based on the arrival time prediction results and the passenger aggregation curve of the station. Based on the route travel time and congestion status, the dispatch cost of fuel consumption, driving time, and off-peak empty driving in the dispatch data in the solution pool is predicted to generate the dispatch cost prediction result; The bus dispatch optimization is performed under the passenger waiting time target and the dispatch cost target according to the average waiting time and dispatch cost prediction results.

5. The intelligent public transportation combined dispatching method according to claim 4, characterized in that: The bus dispatch optimization under the passenger waiting time target and the dispatch cost target is performed according to the average waiting time and the dispatch cost prediction result, including: Get the maximum capacity of the bus; Carry out vehicle passenger density prediction for the dispatching data in the solution pool and establish the passenger density prediction result; Calculating an effective passenger rate based on the maximum capacity and passenger density prediction results to generate an effective passenger rate calculation result; Obtaining the arrival time of the congested road section according to the scheduling data in the solution pool, using the arrival time to predict the congestion state, and establishing a basic congestion state; According to the bus length characteristics and stop characteristics, the intervention increment fitting under the basic congestion state is performed to generate the fitted congestion state; The average waiting time, the dispatching cost prediction result, the effective passenger rate calculation result and the fitted congestion state are used to optimize bus dispatch under the passenger waiting time target, the dispatching cost target, the vehicle utilization rate target and the congestion impact target.

6. The intelligent public transportation combined dispatching method according to claim 1, characterized in that: The bus dispatch management according to the optimization result includes: Continue to add matching records triggered by navigation sites and build incremental site data; The incremental data of the stations are used to fine-tune the scheduling of the optimization results to complete the bus scheduling management.

7. The intelligent public transportation combined dispatching method according to claim 1, characterized in that: The step of establishing a spatiotemporal heat prediction map includes: Identify the main passengers and establish specific passenger groups; The scheduling preference of a specific passenger group is set, and the scheduling preference is added to the spatiotemporal heat prediction map.

8. A smart bus combined dispatching device, characterized in that: The device is used to implement the intelligent bus combined dispatching method according to any one of claims 1 to 7, comprising: Information calling module, used to call bus card swiping records and establish a card swiping site feature set; A matching analysis module is used to establish a navigation site trigger, use the navigation site trigger to perform navigation record matching analysis, and establish a spatiotemporal heat prediction map based on the matching analysis results and the card swiping site feature set; A state prediction module is used to construct a multi-source traffic flow dataset, use the multi-source traffic flow dataset to predict road congestion status, and construct a regional road congestion status map; An information flow module is used to establish station passenger flow demand, route travel time and congestion status based on the spatiotemporal heat prediction map and the regional road congestion status map; A channel configuration module is used to configure a multi-objective scheduling channel, taking the passenger flow demand of the station, the route travel time and the congestion status as input data, and optimizing the route bus scheduling based on the passenger waiting time target, the scheduling cost target, the vehicle utilization rate target and the congestion impact target, and establishing the optimization result; The dispatching management module is used to perform public transportation dispatching management according to the optimization result.

9. A computer device, characterized in that: The computer device comprises: a memory for storing executable instructions; The processor is configured to implement the intelligent bus combination scheduling method described in any one of claims 1 to 7 when running the executable instructions stored in the memory.

10. A non-transitory computer-readable storage medium, characterized in that The storage medium stores a computer software program, which, when executed by a processor, implements a smart bus combination scheduling method as described in any one of claims 1 to 7.

Images