Sanitation equipment scheduling method and system based on Internet of Things

Abstract

The invention discloses a sanitation equipment scheduling method and system based on the Internet of Things, and relates to the technical field of intelligent sanitation, and the method comprises the steps: monitoring the real-time garbage quality of each garbage station through the Internet of Things, and automatically generating a clearance order containing the position, the estimated garbage quality and the priority score when the real-time garbage quality exceeds a threshold value; the transport capacity state of the sanitation vehicle is obtained in real time, and the arrival time of the vehicle to the order position is predicted in combination with traffic data and historical information; calculating a matching income value of the vehicle and the order, wherein the matching income value integrates time efficiency, load matching and order priority; based on the matching income value, solving an optimal allocation scheme through integer programming under the condition that the vehicle residual load constraint is met, and generating an order dispatching instruction; the system comprises an order generation module, a vehicle information and time prediction module, a matching income calculation module and an order distribution module. Intelligent scheduling of the environmental sanitation equipment is realized, and the clearing efficiency and the resource utilization rate are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent sanitation technology, specifically to a method and system for scheduling sanitation equipment based on the Internet of Things. Background Technology

[0002] With the acceleration of urbanization, the built-up area of ​​cities is constantly expanding and the population density is continuously increasing, resulting in a year-on-year increase in the amount of domestic waste generated. This places higher demands on the timeliness, efficiency, and resource optimization capabilities of sanitation and waste collection. The traditional sanitation equipment dispatching model, which is based on "manual inspection + experience-based dispatching," is gradually becoming inadequate for the sanitation operation needs in the context of smart cities. Its limitations are mainly reflected in the following aspects: Traditional methods rely on staff to periodically inspect the filling status of waste stations, with inspection cycles typically ranging from several hours to a day, making it impossible to dynamically capture real-time changes in waste volume. If the inspection intervals are too long, environmental pollution problems such as excessive waste accumulation, odor spread, or sewage leakage can easily occur. Shortening the inspection cycle to avoid overflow significantly increases labor costs and still fails to cover the real-time status of all stations. While some scenarios introduce simple sensors to monitor whether waste stations are "overflowing," they only output a binary "full / not full" signal, unable to combine historical collection data to estimate the total waste volume, making it difficult to accurately match transport capacity when dispatching vehicles.

[0003] Current sanitation vehicle dispatching strategies largely employ "nearby dispatch" or "fixed-route cyclical collection," failing to fully integrate multi-source key data. On one hand, they don't link vehicle capacity status with order demand in real time, often resulting in insufficient or redundant load capacity after vehicles arrive at garbage stations. On the other hand, they don't consider order urgency and real-time traffic conditions, potentially delaying urgent orders or causing vehicles to take excessively long detours due to congestion, reducing overall collection efficiency. Furthermore, dispatching decisions rely on manual experience, making it impossible to globally optimize combinations of multiple vehicles and orders, leading to persistently high empty-running rates for sanitation vehicles and significant resource waste.

[0004] In traditional dispatching systems, waste station monitoring data, vehicle operation data, traffic data, and historical operation data are isolated from each other and lack a collaborative analysis mechanism. For example, historical waste collection data is not used to optimize prediction models, and historical vehicle driving data is not used to improve the accuracy of arrival time predictions. This results in dispatching schemes lacking data support for scientific rigor and foresight, making it difficult to balance the dual requirements of "timely collection" and "resource utilization."

[0005] In summary, the current sanitation equipment dispatching faces core pain points such as "insufficient real-time monitoring, imbalance in transport capacity matching, and low decision-making efficiency." A solution based on Internet of Things (IoT) technology is needed to integrate multi-source data and achieve dynamic and precise dispatching, so as to improve sanitation operation efficiency, reduce operating costs, and meet the sanitation management needs of modern smart cities. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for scheduling sanitation equipment based on the Internet of Things (IoT) to solve the problems raised in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method and system for scheduling sanitation equipment based on the Internet of Things (IoT). The methods include: Step S1: Monitor the real-time waste quality data of each waste station. When the waste quality of a certain waste station exceeds a preset threshold, automatically generate a collection order. Each collection order includes at least the order location information, the estimated waste quality calculated based on historical data and real-time waste quality, and a priority score that represents the urgency of collection. Step S2: Obtain the real-time transport capacity status information of each sanitation vehicle, including the vehicle's current location, real-time load, and remaining load; and based on real-time traffic data, predict the arrival time of each sanitation vehicle to the location of each of the aforementioned collection orders. Step S3: For each pairing of sanitation vehicle and collection order, calculate a matching revenue value; Step S4: Based on the matching revenue value, under the constraint that the total estimated garbage weight of a single vehicle allocation order does not exceed its remaining load, sanitation vehicles are allocated to each collection order, and a dispatch instruction is generated.

[0008] Furthermore, step S1 includes: Step S1-1: Collect the training dataset of the garbage station. The training dataset includes the historical garbage quality data of the garbage station at the time of each collection event and the actual garbage quality recorded in the corresponding collection event. Input the training dataset into the machine learning algorithm for training to establish a garbage quality prediction model f with the historical garbage quality data as input and the actual garbage quality as output target. Step S1-2: Obtain real-time waste quality data for each waste station according to a preset period T. The real-time waste quality of a certain waste station i at time t is... Its preset waste quality threshold is The real-time waste quality data With waste quality threshold Perform a comparison; when At that time, the process of generating a collection order for garbage station i is triggered; Steps S1-3: Determine the order location based on the preset geographical location information of garbage station i. Call the prediction model f, and input the historical collection volume dataset of garbage station i. With real-time waste quality The estimated waste mass was calculated. ; Step S1-4: Obtain historical waste quality data for waste station i over n periods, and calculate the sequence of changes in waste quality between adjacent periods. The calculation formula is as follows: Where k is an integer from 1 to n-1, For the garbage station i, push forward at time t Real-time waste quality for each cycle, For the garbage station i, push forward at time t Real-time waste quality for each cycle; calculate the waste quality growth rate based on the change sequence. The calculation formula is: ; Steps S1-5: Based on the waste quality growth rate Calculate priority score The calculation formula is: ,in and The preset weighting coefficients and ; Steps S1-6: Set the order location Estimate the quality of waste and priority score Bind the data and generate a waste disposal order m.

[0009] Furthermore, step S2 includes: Step S2-1: Obtain the current location of sanitation vehicle j in real time using the global positioning system installed on the vehicle. Simultaneously, the vehicle's load status information is acquired in real time, including its real-time load. and maximum load capacity And calculate the vehicle's remaining load capacity using the following formula: ; Step S2-2: For each pair of sanitation vehicles j and collection orders m, based on the current location of vehicle j... and the order location of the cleaning order m Using the route planning algorithm provided by the electronic map service, calculate from arrive The distance of the optimal driving path ; Step S2-3: Based on path distance Average operating speed of sanitation vehicles Calculate the baseline travel time The calculation formula is: The average operating speed This is a constant derived from historical driving data statistics; Step S2-4: Obtain N historical collection and transportation task data for sanitation vehicle j, wherein the historical collection and transportation task data includes the path distance of the historical collection and transportation tasks. and actual time spent ; Calculate the historical average baseline travel time of sanitation vehicle j. Compared with historical average actual time The calculation formulas are as follows: and Finally, a vehicle performance factor characterizing its reliability is calculated. The calculation formula is: ; Step S2-5: Calculate the final predicted arrival time The calculation formula is: .

[0010] Furthermore, step S3 includes: Step S3-1: Filter to meet the requirements Validally match sanitation vehicle j with collection order m, and obtain the predicted arrival time for each valid match. Estimated waste quality for orders Order priority score and the remaining load of vehicle j ; Step S3-2: Calculate the time efficiency factor of the effective pairing. and load matching step factor The time efficiency factor The calculation formula is: The load matching factor The calculation formula is: Where min is the function for finding the minimum value, and max is the function for finding the maximum value; Step S3-3: Based on time efficiency factor Load matching factor Order priority score The matching payoff value is calculated by weighted summation. : ,in, The preset weighting coefficients and .

[0011] Furthermore, step S4 includes: Step S4-1: Construct an allocation optimization model with the objective of maximizing the total matching revenue of the system. The objective function of the optimization model is: Where J represents the total number of sanitation vehicles, and M represents the total number of waste collection orders to be assigned. Let be the matching revenue value between vehicle j and order m. For binary decision variables, when vehicle j is assigned to order m, ,otherwise ; Step S4-2: Set the following constraints for the optimization model: Each waste collection order must be completed and can only be served by one sanitation vehicle, that is: The sum of the estimated garbage weights for all orders assigned to the same sanitation vehicle must not exceed the vehicle's remaining load capacity, i.e.: ; Step S4-3: Use an integer programming solver to solve the optimization model constructed in step S4-1 and constrained by step S4-2 to obtain the optimal allocation scheme; Step S4-4: According to the optimal allocation scheme, issue a dispatch instruction to each sanitation vehicle j that has been assigned a task; the dispatch instruction shall include at least: a list of all cleaning orders assigned to the vehicle, a suggested cleaning sequence generated based on the order location information, and the optimal driving route to each order location calculated by the route planning service.

[0012] The system includes: an order generation module, a vehicle information and time prediction module, a matching revenue calculation module, and an order allocation module; The order generation module is used to monitor the real-time waste quality data of each waste station. When the waste quality of a certain waste station exceeds a preset threshold, a collection order is automatically generated. Each collection order includes at least the order location information, the estimated waste quality calculated based on historical data and real-time waste quality, and a priority score that represents the urgency of collection. The vehicle information and time prediction module is used to obtain the transportation status information of each sanitation vehicle in real time. The transportation status information includes the vehicle's current location, real-time load, and remaining load. Based on real-time traffic data, it predicts the arrival time of each sanitation vehicle to the location of each of the collection orders. The matching revenue calculation module is used to calculate a matching revenue value for each sanitation vehicle and collection order. The order allocation module is used to allocate sanitation vehicles to each collection order based on the matching revenue value, under the constraint that the total estimated garbage weight of a single vehicle allocation order does not exceed its remaining load, and to generate a dispatch instruction.

[0013] The order generation module includes: The prediction model training unit is used to collect the training dataset of the garbage station. The training dataset includes the historical garbage quality data of the garbage station at the time of each collection event and the actual garbage quality recorded in the corresponding collection event. The training dataset is input into the machine learning algorithm for training to establish a predicted garbage quality model f with the historical garbage quality data as input and the actual garbage quality as the output target. The waste quality monitoring and triggering unit is used to acquire real-time waste quality data of each waste station according to a preset period T. The real-time waste quality of a certain waste station i at time t is: Its preset waste quality threshold is The real-time waste quality data With waste quality threshold Perform a comparison; when At that time, the process of generating a collection order for garbage station i is triggered; The estimated quality calculation unit is used to determine the order location based on the preset geographical location information of waste station i. Call the prediction model f, and input the historical collection volume dataset of garbage station i. With real-time waste quality The estimated waste mass was calculated. ; The growth rate calculation unit is used to acquire historical waste quality data of waste station i over n periods and calculate the sequence of changes in waste quality between adjacent periods. The calculation formula is as follows: Where k is an integer from 1 to n-1, Let i be the real-time waste quality of waste station i, k cycles prior to time t. The real-time waste quality of waste station i at time t, k+1 periods prior; based on the change sequence, the waste quality growth rate is calculated. The calculation formula is: ; Priority calculation unit, used for prioritization based on waste quality growth rate Calculate priority score The calculation formula is: in and The preset weighting coefficients and ; The order binding generation unit is used to bind the order location. Estimate the quality of waste and priority score Bind the data and generate a waste disposal order m.

[0014] The vehicle information and time prediction module includes: The vehicle status acquisition unit is used to acquire the current location of the sanitation vehicle j in real time through a global positioning system installed on the vehicle. Simultaneously, the vehicle's load status information is acquired in real time, including its real-time load. and maximum load capacity And calculate the vehicle's remaining load capacity using the following formula: ; The path distance calculation unit is used to calculate the distance between each pair of sanitation vehicles j and collection orders m, based on the current location of vehicle j. and the order location of the cleaning order m Using the route planning algorithm provided by the electronic map service, calculate from arrive The distance of the optimal driving path ; The base time calculation unit is used to calculate the time based on the path distance. Average operating speed of sanitation vehicles Calculate the baseline travel time The calculation formula is: The average operating speed This is a constant derived from historical driving data statistics; The vehicle efficiency factor calculation unit is used to obtain N historical collection and transportation task data of sanitation vehicle j, wherein the historical collection and transportation task data includes the path distance of the historical collection and transportation tasks. and actual time spent ; Calculate the historical average baseline travel time of sanitation vehicle j. Compared with historical average actual time The calculation formulas are as follows: Finally, a vehicle performance factor characterizing its reliability is calculated. The calculation formula is: ; The predicted arrival time calculation unit is used to calculate the final predicted arrival time. The calculation formula is: .

[0015] The matching profit calculation module includes: Effective paired filter units are used to filter those that meet the following criteria. Validally match sanitation vehicle j with collection order m, and obtain the predicted arrival time for each valid match. Estimated waste quality for orders Order priority score and the remaining load of vehicle j ; The efficiency factor calculation unit is used to calculate the time efficiency factor of the effective pairing. and load matching factor The time efficiency factor The calculation formula is: The load matching factor The calculation formula is: Where min is the function for finding the minimum value, and max is the function for finding the maximum value; Matching return value calculation unit, used for time efficiency factor Load matching factor Order priority score The matching payoff value is calculated by weighted summation. : ,in, The preset weighting coefficients and .

[0016] The order allocation module includes: An optimization model building unit is used to construct an allocation optimization model with the objective of maximizing the total matching revenue of the system. The objective function of the optimization model is: Where J represents the total number of sanitation vehicles, and M represents the total number of waste collection orders to be assigned. Let be the matching revenue value between vehicle j and order m. For binary decision variables, when vehicle j is assigned to order m, =1, otherwise ; The constraint setting unit is used to set the following constraints for the optimization model: each collection order must be completed and can only be served by one sanitation vehicle, that is: The sum of the estimated garbage weights for all orders assigned to the same sanitation vehicle must not exceed the vehicle's remaining load capacity, i.e.: ; The model solving unit is used to solve the optimization model constructed by the optimization model building unit and constrained by the constraint setting unit using an integer programming solver, so as to obtain the optimal allocation scheme. The dispatch instruction generation unit is used to issue dispatch instructions to each sanitation vehicle j assigned a task according to the optimal allocation scheme; the dispatch instruction includes at least: a list of all cleaning orders assigned to the vehicle, a suggested cleaning sequence generated based on the order location information, and the optimal driving route to each order location calculated by the route planning service.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. An optimization model that maximizes the overall matching revenue of the system is constructed using integer programming. Under the constraints of load capacity and the requirement of guaranteed order delivery, a global solution is performed for combinations of multiple vehicles and multiple orders. Compared with traditional manual experience or fixed route scheduling, this solution can break the limitations of local optima, significantly reduce the empty running rate of vehicles, and increase the average daily collection volume of a single vehicle. It balances "timeliness" and "resource efficiency" at the global level, directly reducing operating costs.

[0018] 2. Innovatively, it weights and quantifies three core factors—time efficiency, load matching, and order priority—into a matching benefit value, replacing the traditional extensive "nearby order dispatch" model. This ensures timely processing of high-priority orders while avoiding issues such as insufficient or redundant vehicle load, thus improving the efficiency of matching orders with individual vehicles.

[0019] 3. Based on real-time collection of waste quality data through the Internet of Things, and combined with a predictive model trained on historical data, the system accurately estimates the collection volume, replacing the traditional binary judgment of "full / not full" based on manual inspections or simple sensors. This not only avoids pollution such as odors and leakage caused by excessive waste accumulation, but also provides accurate data support for subsequent capacity matching, solving the problem of "capacity mismatch caused by prediction errors" in traditional scheduling. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a sanitation equipment dispatching system based on the Internet of Things according to the present invention; Figure 2 This is a schematic diagram of the vehicle information and time prediction module of an Internet of Things-based sanitation equipment dispatching system according to the present invention. Figure 3 This is a flowchart illustrating a sanitation equipment scheduling method based on the Internet of Things according to the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example: Figure 1 , Figure 2 and Figure 3 As shown, the present invention provides a technical solution: a method and system for scheduling sanitation equipment based on the Internet of Things. The methods include: Step S1: Monitor the real-time waste quality data of each waste station. When the waste quality of a certain waste station exceeds a preset threshold, automatically generate a collection order. Each collection order includes at least the order location information, the estimated waste quality calculated based on historical data and real-time waste quality, and a priority score that represents the urgency of collection. Step S2: Obtain the real-time transport capacity status information of each sanitation vehicle, including the vehicle's current location, real-time load, and remaining load; and based on real-time traffic data, predict the arrival time of each sanitation vehicle to the location of each of the aforementioned collection orders. Step S3: For each pairing of sanitation vehicle and collection order, calculate a matching revenue value; Step S4: Based on the matching revenue value, under the constraint that the total estimated garbage weight of a single vehicle allocation order does not exceed its remaining load, sanitation vehicles are allocated to each collection order, and a dispatch instruction is generated.

[0023] Furthermore, step S1 includes: Step S1-1: The system collects historical training datasets from the waste station, including historical waste quality data at the time of each collection event, and the actual waste quality recorded in the corresponding collection event. The training dataset is then input into a gradient boosting decision tree machine learning algorithm for training, establishing a waste quality prediction model f that uses historical waste quality data as input and actual waste quality as the output target. Step S1-2: The system acquires real-time waste quality data for each waste station at a preset period of T=30min; for waste station i1, the real-time waste quality at time t=10:00 is... Its preset waste quality threshold is ;because This triggers the process of generating a waste collection order for waste station i1; the unit for waste quality is t (tons); Steps S1-3: Determine the order location based on the preset geographical location information of garbage station i1. Call the prediction model f, and input the historical collection volume dataset of garbage station i1. and real-time waste quality The estimated waste quality was calculated. ; Steps S1-4: Obtain historical waste quality data for waste station i1 over n=6 periods: , ; Calculate the sequence of changes in waste quality between adjacent periods: , , , , ; Calculate the waste quality growth rate based on the change series. : ; Step S1-5: Set weighting coefficients , Based on the waste quality growth rate Calculate priority score : ; Steps S1-6: Bind order location Estimate the quality of waste Priority Order001 is generated.

[0024] At the same time, the garbage station i2 also triggers order generation: order location Estimate the quality of waste Priority Generate a waste disposal order, Order002.

[0025] Furthermore, step S2 includes: Step S2-1: Obtain the current location of sanitation vehicle j in real time using the global positioning system installed on the vehicle. At the same time, the vehicle's load status information is obtained in real time; Vehicle J1 Status: Current Location Real-time load Maximum load Remaining load capacity ; Vehicle J2 Status: Current Location Real-time load Maximum load Remaining load capacity ; Step S2-2: For each pair of sanitation vehicles j and collection orders m, based on the current location of the vehicles... and the location of the cleaning order The optimal driving route distance is calculated using the route planning algorithm provided by the electronic map service. ; ; Step S2-3: Based on path distance Average operating speed of sanitation vehicles km / h, calculate the baseline travel time: ; ; ; ; Step S2-4: Obtain N historical collection and transportation task data for sanitation vehicle j, including the path distance of the historical collection and transportation tasks. and actual time spent ; Vehicle J1 historical data (N=50 times): Historical average baseline travel time: Historical average actual time: Vehicle efficiency factor: ; Vehicle J2 historical data (N=45 times): Historical average baseline travel time: Historical average actual time: Vehicle efficiency factor: ;

[0026] Step S2-5: Calculate the final predicted travel time : ; ; ; ; Furthermore, step S3 includes: Step S3-1: Filter to meet the requirements Valid pairings of sanitation vehicle j with collection order m are determined; all pairings satisfy the conditions: Order001 (1.25t) ≤ j1 (3.5t), j2 (4.0t); Order002 (0.85t) ≤ j1 (3.5t), j2 (4.0t); predict the arrival time for each valid pairing is obtained. Estimated waste quality for orders Order priority score and the remaining load of vehicle j ; Step S3-2: Calculate the time efficiency factor and load matching factor : Time efficiency factor : ; ; ; ; Load matching factor : ; ; ; ; Step S3-3: Based on time efficiency factor Load matching factor Order priority score The matching payoff value is calculated by weighted summation. ,in : ; ; ; .

[0027] Furthermore, step S4 includes: Step S4-1: Construct an allocation optimization model with the objective of maximizing the total matching revenue of the system. The objective function is: Where J=2 is the total number of sanitation vehicles, and M=2 is the total number of waste collection orders to be assigned; Specific expression: ; Step S4-2: Set constraints for the optimization model: Each waste collection order must be completed and can only be served by one sanitation vehicle: For Order001: ; For Order002: ; The sum of the estimated garbage weights for all orders assigned to the same sanitation vehicle must not exceed the vehicle's remaining load capacity. For j1:

[0028] For j2:

[0029] Step S4-3: Solve the optimization model using an integer programming solver to obtain the optimal allocation scheme: j2 assigns Order001:

[0030] j1 assigns Order002:

[0031] Total matching profit = 3.670 + 1.645 = 5.315; Step S4-4: Based on the optimal allocation scheme, issue a dispatch instruction to each sanitation vehicle assigned a task: Vehicle J2 dispatch instruction: Assigning cleaning order: Order001; Order location: (116.4075°, 39.9042°); Recommended disposal sequence: Proceed directly to garbage station i1; Optimal driving route: distance 3.1km, estimated time 8.5min; Estimated waste weight to be removed: 1.25 tons; Vehicle J1 dispatch instruction: Assigning cleaning and disposal order: Order002; Order location: (116.4164°, 39.9288°); Recommended disposal sequence: Proceed directly to garbage station i2; Optimal driving route: 6.8km, estimated time 19.6min; Estimated waste weight to be removed: 0.85 tons.

[0032] The system includes: an order generation module, a vehicle information and time prediction module, a matching revenue calculation module, and an order allocation module; The order generation module is used to monitor the real-time waste quality data of each waste station. When the waste quality of a certain waste station exceeds a preset threshold, a collection order is automatically generated. Each collection order includes at least the order location information, the estimated waste quality calculated based on historical data and real-time waste quality, and a priority score that represents the urgency of collection. The vehicle information and time prediction module is used to obtain the transportation status information of each sanitation vehicle in real time. The transportation status information includes the vehicle's current location, real-time load, and remaining load. Based on real-time traffic data, it predicts the arrival time of each sanitation vehicle to the location of each of the collection orders. The matching revenue calculation module is used to calculate a matching revenue value for each sanitation vehicle and collection order. The order allocation module is used to allocate sanitation vehicles to each collection order based on the matching revenue value, under the constraint that the total estimated garbage weight of a single vehicle allocation order does not exceed its remaining load, and to generate a dispatch instruction.

[0033] The order generation module includes: The prediction model training unit is used to collect the training dataset of the garbage station. The training dataset includes the historical garbage quality data of the garbage station at the time of each collection event and the actual garbage quality recorded in the corresponding collection event. The training dataset is input into the machine learning algorithm for training to establish a predicted garbage quality model f with the historical garbage quality data as input and the actual garbage quality as the output target. The waste quality monitoring and triggering unit is used to acquire real-time waste quality data of each waste station according to a preset period T. The real-time waste quality of a certain waste station i at time t is: Its preset waste quality threshold is The real-time waste quality data With waste quality threshold Perform a comparison; when At that time, the process of generating a collection order for garbage station i is triggered; The estimated quality calculation unit is used to determine the order location based on the preset geographical location information of waste station i. Call the prediction model f, and input the historical collection volume dataset of garbage station i. With real-time waste quality The estimated waste mass was calculated. ; The growth rate calculation unit is used to acquire historical waste quality data of waste station i over n periods and calculate the sequence of changes in waste quality between adjacent periods. The calculation formula is as follows: Where k is an integer from 1 to n-1, Let i be the real-time waste quality of waste station i, k cycles prior to time t. The real-time waste quality of waste station i at time t, k+1 periods prior; based on the change sequence, the waste quality growth rate is calculated. The calculation formula is: ; Priority calculation unit, used for prioritization based on waste quality growth rate Calculate priority score The calculation formula is: ,in and The preset weighting coefficients and ; The order binding generation unit is used to bind the order location. Estimate the quality of waste and priority score Bind the data and generate a waste disposal order m.

[0034] The vehicle information and time prediction module includes: The vehicle status acquisition unit is used to acquire the current location of the sanitation vehicle j in real time through a global positioning system installed on the vehicle. Simultaneously, the vehicle's load status information is acquired in real time, including its real-time load. and maximum load capacity And calculate the vehicle's remaining load capacity using the following formula: ; The path distance calculation unit is used to calculate the distance between each pair of sanitation vehicles j and collection orders m, based on the current location of vehicle j. and the order location of the cleaning order m Using the route planning algorithm provided by the electronic map service, calculate from arrive The distance of the optimal driving path ; The base time calculation unit is used to calculate the time based on the path distance. Average operating speed of sanitation vehicles Calculate the baseline travel time The calculation formula is: The average operating speed This is a constant derived from historical driving data statistics; The vehicle efficiency factor calculation unit is used to obtain N historical collection and transportation task data of sanitation vehicle j, wherein the historical collection and transportation task data includes the path distance of the historical collection and transportation tasks. and actual time spent ; Calculate the historical average baseline travel time of sanitation vehicle j. Compared with historical average actual time The calculation formulas are as follows: Finally, a vehicle performance factor characterizing its reliability is calculated. The calculation formula is: ; The predicted arrival time calculation unit is used to calculate the final predicted arrival time. The calculation formula is: .

[0035] The matching profit calculation module includes: Effective paired filter units are used to filter those that meet the following criteria. Validally match sanitation vehicle j with collection order m, and obtain the predicted arrival time for each valid match. Estimated waste quality for orders Order priority score and the remaining load of vehicle j ; The efficiency factor calculation unit is used to calculate the time efficiency factor of the effective pairing. and load matching factor The time efficiency factor The calculation formula is: The load matching factor The calculation formula is: Where min is the function for finding the minimum value, and max is the function for finding the maximum value; Matching return value calculation unit, used for time efficiency factor Load matching factor Order priority score The matching payoff value is calculated by weighted summation. ,in, The preset weighting coefficients and .

[0036] The order allocation module includes: An optimization model building unit is used to construct an allocation optimization model with the objective of maximizing the total matching revenue of the system. The objective function of the optimization model is: Where J represents the total number of sanitation vehicles, and M represents the total number of waste collection orders to be assigned. Let be the matching revenue value between vehicle j and order m. For binary decision variables, when vehicle j is assigned to order m, ,otherwise ; The constraint setting unit is used to set the following constraints for the optimization model: each collection order must be completed and can only be served by one sanitation vehicle, that is: The sum of the estimated garbage weights for all orders assigned to the same sanitation vehicle must not exceed the vehicle's remaining load capacity, i.e.: ; The model solving unit is used to solve the optimization model constructed by the optimization model building unit and constrained by the constraint setting unit using an integer programming solver, so as to obtain the optimal allocation scheme. The dispatch instruction generation unit is used to issue dispatch instructions to each sanitation vehicle j assigned a task according to the optimal allocation scheme; the dispatch instruction includes at least: a list of all cleaning orders assigned to the vehicle, a suggested cleaning sequence generated based on the order location information, and the optimal driving route to each order location calculated by the route planning service.

[0037] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for scheduling sanitation equipment based on the Internet of Things, characterized in that: Includes the following steps: Step S1: Monitor the real-time waste quality data of each waste station. When the waste quality of a certain waste station exceeds a preset threshold, automatically generate a collection order. Each collection order includes at least the order location information, the estimated waste quality calculated based on historical data and real-time waste quality, and a priority score that represents the urgency of collection. Step S2: Obtain the real-time transport capacity status information of each sanitation vehicle, including the vehicle's current location, real-time load, and remaining load; and based on real-time traffic data, predict the arrival time of each sanitation vehicle to the location of each of the aforementioned collection orders. Step S3: For each pairing of sanitation vehicle and collection order, calculate a matching revenue value; Step S4: Based on the matching revenue value, under the constraint that the total estimated garbage weight of a single vehicle allocation order does not exceed its remaining load, sanitation vehicles are allocated to each collection order, and a dispatch instruction is generated.

2. The method for scheduling sanitation equipment based on the Internet of Things according to claim 1, characterized in that: Step S1 includes: Step S1-1: Collect the training dataset of the garbage station. The training dataset includes the historical garbage quality data of the garbage station at the time of each collection event and the actual garbage quality recorded in the corresponding collection event. Input the training dataset into the machine learning algorithm for training to establish a garbage quality prediction model f with the historical garbage quality data as input and the actual garbage quality as output target. Step S1-2: Obtain real-time waste quality data for each waste station according to a preset period T. The real-time waste quality of a certain waste station i at time t is... Its preset waste quality threshold is The real-time waste quality data With waste quality threshold Perform a comparison; when At that time, the process of generating a collection order for garbage station i is triggered; Steps S1-3: Determine the order location based on the preset geographical location information of garbage station i. Call the prediction model f, and input the historical collection volume dataset of garbage station i. With real-time waste quality The estimated waste mass was calculated. ; Step S1-4: Obtain historical waste quality data for waste station i over n periods, and calculate the sequence of changes in waste quality between adjacent periods. The calculation formula is as follows: Where k is an integer from 1 to n-1, Let i be the real-time waste quality of waste station i, k cycles prior to time t. The real-time waste quality of waste station i at time t, k+1 periods prior; based on the change sequence, the waste quality growth rate is calculated. The calculation formula is: ; Steps S1-5: Based on the waste quality growth rate Calculate priority score The calculation formula is: in and The preset weighting coefficients and ; Steps S1-6: Set the order location Estimate the quality of waste and priority score Bind the data and generate a waste disposal order m.

3. The method for scheduling sanitation equipment based on the Internet of Things according to claim 2, characterized in that: Step S2 includes: Step S2-1: Obtain the current location of sanitation vehicle j in real time using the global positioning system installed on the vehicle. Simultaneously, the vehicle's load status information is acquired in real time, including its real-time load. and maximum load capacity And calculate the vehicle's remaining load capacity using the following formula: ; Step S2-2: For each pair of sanitation vehicles j and collection orders m, based on the current location of vehicle j... and the order location of the cleaning order m Using the route planning algorithm provided by the electronic map service, calculate from arrive The distance of the optimal driving path ; Step S2-3: Based on path distance Average operating speed of sanitation vehicles Calculate the baseline travel time The calculation formula is: The average operating speed This is a constant derived from historical driving data statistics; Step S2-4: Obtain N historical collection and transportation task data for sanitation vehicle j, wherein the historical collection and transportation task data includes the path distance of the historical collection and transportation tasks. and actual time spent ; Calculate the historical average baseline travel time of sanitation vehicle j. Compared with historical average actual time The calculation formulas are as follows: and Finally, a vehicle performance factor characterizing its reliability is calculated. The calculation formula is: ; Step S2-5: Calculate the final predicted arrival time The calculation formula is: .

4. The method for scheduling sanitation equipment based on the Internet of Things according to claim 3, characterized in that: Step S3 includes: Step S3-1: Filter to meet the requirements Validally match sanitation vehicle j with collection order m, and obtain the predicted arrival time for each valid match. Estimated waste quality for orders Order priority score and the remaining load of vehicle j ; Step S3-2: Calculate the time efficiency factor of the effective pairing. and load matching factor The time efficiency factor The calculation formula is: The load matching factor The calculation formula is: Where min is the function for finding the minimum value, and max is the function for finding the maximum value; Step S3-3: Based on time efficiency factor Load matching factor Order priority score The matching payoff value is calculated by weighted summation: ,in, , is the preset weighting coefficient and .

5. The sanitation equipment scheduling method based on the Internet of Things according to claim 4, characterized in that: Step S4 includes: Step S4-1: Construct an allocation optimization model with the objective of maximizing the total matching revenue of the system. The objective function of the optimization model is: Where J represents the total number of sanitation vehicles, and M represents the total number of waste collection orders to be assigned. Let be the matching revenue value between vehicle j and order m. For binary decision variables, when vehicle j is assigned to order m, ,otherwise ; Step S4-2: Set the following constraints for the optimization model: Each waste collection order must be completed and can only be served by one sanitation vehicle, that is: The sum of the estimated garbage weights for all orders assigned to the same sanitation vehicle must not exceed the vehicle's remaining load capacity, i.e.: ; Step S4-3: Use an integer programming solver to solve the optimization model constructed in step S4-1 and constrained by step S4-2 to obtain the optimal allocation scheme; Step S4-4: According to the optimal allocation scheme, issue a dispatch instruction to each sanitation vehicle j that has been assigned a task; the dispatch instruction shall include at least: a list of all cleaning orders assigned to the vehicle, a suggested cleaning sequence generated based on the order location information, and the optimal driving route to each order location calculated by the route planning service.

6. An Internet of Things (IoT)-based sanitation equipment scheduling system, used to execute the IoT-based sanitation equipment scheduling method according to any one of claims 1-5, characterized in that: The system includes: The module includes an order generation module, a vehicle information and time prediction module, a matching revenue calculation module, and an order allocation module. The order generation module is used to monitor the real-time waste quality data of each waste station. When the waste quality of a certain waste station exceeds a preset threshold, a collection order is automatically generated. Each collection order includes at least the order location information, the estimated waste quality calculated based on historical data and real-time waste quality, and a priority score that represents the urgency of collection. The vehicle information and time prediction module is used to obtain the transportation status information of each sanitation vehicle in real time. The transportation status information includes the vehicle's current location, real-time load, and remaining load. Based on real-time traffic data, it predicts the arrival time of each sanitation vehicle to the location of each of the collection orders. The matching revenue calculation module is used to calculate a matching revenue value for each sanitation vehicle and collection order. The order allocation module is used to allocate sanitation vehicles to each collection order based on the matching revenue value, under the constraint that the total estimated garbage weight of a single vehicle allocation order does not exceed its remaining load, and to generate a dispatch instruction.

7. The method for scheduling sanitation equipment based on the Internet of Things according to claim 6, characterized in that: The order generation module includes: The prediction model training unit is used to collect the training dataset of the garbage station. The training dataset includes the historical garbage quality data of the garbage station at the time of each collection event and the actual garbage quality recorded in the corresponding collection event. The training dataset is input into the machine learning algorithm for training to establish a predicted garbage quality model f with the historical garbage quality data as input and the actual garbage quality as the output target. The waste quality monitoring and triggering unit is used to acquire real-time waste quality data of each waste station according to a preset period T. The real-time waste quality of a certain waste station i at time t is: Its preset waste quality threshold is The real-time waste quality data With waste quality threshold Perform a comparison; when At that time, the process of generating a collection order for garbage station i is triggered; The estimated quality calculation unit is used to determine the order location based on the preset geographical location information of waste station i. Call the prediction model f, and input the historical collection volume dataset of garbage station i. With real-time waste quality The calculation yielded the estimate. ; The growth rate calculation unit is used to acquire historical waste quality data of waste station i over n periods and calculate the sequence of changes in waste quality between adjacent periods. The calculation formula is as follows: Where k is an integer from 1 to n-1, Let i be the real-time waste quality of waste station i, k cycles prior to time t. The real-time waste quality of waste station i at time t, k+1 periods prior; based on the change sequence, the waste quality growth rate is calculated. The calculation formula is: ; Priority calculation unit, used for prioritization based on waste quality growth rate Calculate priority score The calculation formula is: ,in and The preset weighting coefficients and ; The order binding generation unit is used to bind the order location. Estimate the quality of waste and priority score Bind the data and generate a waste disposal order m.

8. The method for scheduling sanitation equipment based on the Internet of Things according to claim 7, characterized in that: The vehicle information and time prediction module includes: The vehicle status acquisition unit is used to acquire the current location of the sanitation vehicle j in real time through a global positioning system installed on the vehicle. Simultaneously, the vehicle's load status information is acquired in real time, including its real-time load. and maximum load capacity And calculate the vehicle's remaining load capacity using the following formula: ; The path distance calculation unit is used to calculate the distance between each pair of sanitation vehicles j and collection orders m, based on the current location of vehicle j. and the order location of the cleaning order m Using the route planning algorithm provided by the electronic map service, calculate from arrive The distance of the optimal driving path ; The base time calculation unit is used to calculate the time based on the path distance. Average operating speed of sanitation vehicles Calculate the baseline travel time The calculation formula is: The average operating speed This is a constant derived from historical driving data statistics; The vehicle efficiency factor calculation unit is used to obtain N historical collection and transportation task data of sanitation vehicle j, wherein the historical collection and transportation task data includes the path distance of the historical collection and transportation tasks. and actual time spent ; Calculate the historical average baseline travel time of sanitation vehicle j. Compared with historical average actual time The calculation formulas are as follows: and Finally, a vehicle performance factor characterizing its reliability is calculated. The calculation formula is: ; The predicted arrival time calculation unit is used to calculate the final predicted arrival time. The calculation formula is: .

9. A method for scheduling sanitation equipment based on the Internet of Things according to claim 8, characterized in that: The matching benefit calculation module includes: Effective paired filter units are used to filter those that meet the following criteria. Validally match sanitation vehicle j with collection order m, and obtain the predicted arrival time for each valid match. Estimated waste quality for orders Order priority score and the remaining load of vehicle j ; The efficiency factor calculation unit is used to calculate the time efficiency factor of the effective pairing. and load matching factor The time efficiency factor The calculation formula is: The load matching factor The calculation formula is: Where min is the function for finding the minimum value, and max is the function for finding the maximum value; Matching return value calculation unit, used for time efficiency factor Load matching factor Order priority score The matching payoff value is calculated by weighted summation: ,in, The preset weighting coefficients and .

10. A method for scheduling sanitation equipment based on the Internet of Things according to claim 9, characterized in that: The order allocation module includes: An optimization model building unit is used to construct an allocation optimization model with the objective of maximizing the total matching revenue of the system. The objective function of the optimization model is: Where J represents the total number of sanitation vehicles, and M represents the total number of waste collection orders to be assigned. Let be the matching revenue value between vehicle j and order m. For binary decision variables, when vehicle j is assigned to order m, ,otherwise ; The constraint setting unit is used to set the following constraints for the optimization model: each collection order must be completed and can only be served by one sanitation vehicle, that is: The sum of the estimated garbage weights for all orders assigned to the same sanitation vehicle must not exceed the vehicle's remaining load capacity, i.e.: ; The model solving unit is used to solve the optimization model constructed by the optimization model building unit and constrained by the constraint setting unit using an integer programming solver, so as to obtain the optimal allocation scheme. The dispatch instruction generation unit is used to issue dispatch instructions to each sanitation vehicle j assigned a task according to the optimal allocation scheme; the dispatch instruction includes at least: a list of all cleaning orders assigned to the vehicle, a suggested cleaning sequence generated based on the order location information, and the optimal driving route to each order location calculated by the route planning service.

Images