Interworking table optimization analysis processing method and system for different service operation modes

Abstract

The application discloses a method and system for optimizing and analyzing a duty table for different service operation modes, and particularly relates to the fields of rail transit crew duty arrangement and computer data processing, which comprises reading a target line, a timetable purpose and an operation mode, searching a scene identifier, a line timetable and rule parameters, and writing associated scene configuration records; reading the line timetable according to the scene configuration records, converting the arrival time or departure time into absolute time combined with the next day mark, and merging adjacent train numbers according to the train number connection relationship and splicing rules in the scene configuration records; establishing scene configuration records according to the operation mode, analyzing special train numbers in the timetable into to-be-arranged train number records, calculating train connection relationships to form a candidate connection diagram, completing duty arrangement and checking under the constraints of value-riding, rest and attendance, and outputting the optimized and analyzed result of the duty table.

Description

Technical Field

[0001] This invention relates to the field of rail transit crew route scheduling and computer data processing technology, and more specifically, to a route table optimization and parsing processing method and system for different business operation modes. Background Technology

[0002] In the compilation of rail transit crew routes, the existing system usually connects train numbers according to arrival and departure times and station relationships based on lines, stations, sections, timetables, and rules for departure, arrival, transfer, shift change and rest, in order to generate driver duty routes. Taking a subway line with mixed routes during morning, noon and evening peak hours as an example, the weekday timetable contains cross-day trains, trains returning to the depot, peak-hour standby trains and turnaround trains. Under the condition of not changing the established timetable and not affecting the continuous operation of trains, it is necessary to ensure that each train to be scheduled is only assigned to one route, the connection point and connection time of the trains meet the requirements, the continuous duty time does not exceed 2 hours, the continuous duty interval is not less than 15 minutes, and the rules for departure, departure, shift change and stay-at-home under different scenarios are met. Existing processing methods often directly overlay various business rules into fixed parameter configurations. The cross-day, turnaround, and special trains in the original timetable do not first form a unified relationship for the trains to be scheduled. After the operation mode or parameter combination changes, the same train may be repeatedly connected, not connected, or have inconsistent rest information after connection under different rules. The resulting routes still need to be manually split and adjusted one by one. The technical problem this application aims to solve is: how to perform unified parsing and train connection calculation on timetables containing special train services based on scenario parameters corresponding to different business operation models, and generate a route table that meets the route business rules. Summary of the Invention

[0003] To overcome the aforementioned deficiencies in the prior art, embodiments of the present invention provide a route table optimization and parsing method and system for different business operation modes. By establishing scenario configuration records according to the operation mode, special train numbers in the timetable are parsed into train number records to be scheduled, train number connection relationships are calculated to form a candidate connection diagram, and route arrangement and verification are completed under the constraints of shift work, rest, and departure / arrival, and the route table optimization and parsing results are output to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for optimizing and parsing routing tables for different business operation models, comprising: S1. Read the target route, timetable purpose and operation mode, retrieve the scene identifier, route timetable and rule parameters, and associate and write them into the scene configuration record; S2. Read the route timetable for each train according to the scenario configuration record, convert the arrival time or departure time to the absolute time according to the next day mark, merge adjacent trains according to the train number connection relationship and splicing rules, and write them into the train to be scheduled record. S3. Match the termination position and termination time of the preceding train to be scheduled and the start position and start time of the following train to be scheduled according to the absolute time. Calculate the connection interval and rest duration according to the traffic relationship and write them into the candidate connection edge to form a candidate connection graph. S4. Limit the number of trains to be scheduled to one position before and after each train. Calculate the value of the candidate connecting edge by multiplying the duration, rest duration, and departure / arrival position relationship. Delete candidate connecting edges that violate the scenario configuration record. Use the subsequent trains to be scheduled as ant colony algorithm nodes. Determine the transition probability by multiplying the pheromone quantity by the inverse of the connecting interval. If there are no unused candidate connecting edges, establish a route for the un-arranged trains to be scheduled. Update the pheromone by sorting the number of routes and the discrete results of the duration. Output the initial route arrangement record when the first candidate connecting edge sets of two adjacent rounds are consistent. S5. Based on the initial route schedule record, write the shift type and special train number mark, check the route timetable to determine the number of train number arrangements, compare the termination position, start position and connection interval of adjacent trains, multiply the absolute time check value of the route by the duration and rest duration, and output the route table optimization analysis result.

[0005] In a preferred embodiment, S1 includes: S11. The target route, timetable purpose and operation mode are written in a fixed order to form a scenario identifier. When the source scenario identifier exists, the corresponding parameter configuration record is read and copied. The source scenario identifier in the copied record is replaced with the scenario identifier, and the scenario index record is output. The target route refers to an operating route corresponding to the current route schedule compilation. It is determined by the route name in the route information. The target route in the scenario index record is used to limit the subsequent reading of station information, section information and timetable information to belong to this operating route. The purpose of the timetable is to indicate the applicable date type in the timetable information. The value can be weekday, weekend or special date. It is used to determine the timetable to be used in this route analysis when there are multiple timetables for the same target route. The operation mode refers to the operational organization status of the target route in this route planning. It is determined by shift division, handover location, departure and arrival methods, and train operation organization relationship. It is used to create independent scenario configuration records for the same route under different operational organization statuses. A fixed writing order means writing the target route first, then the timetable purpose, and finally the operation mode, and using the same separator to form a scene identifier; the three components of the scene identifier maintain the same position during creation, copying, and reading. The source scenario identifier refers to the existing scenario identifier that has already formed a scenario configuration record. It is used as the parameter copy object when creating a new scenario. The copy operation only changes the scenario identifier in the parameter configuration record, while the other parameter content and the record writing order remain unchanged. A parameter configuration record is a business parameter record that is bound to a scenario identifier. The applicable location is determined by the site name and direction of the parameter item, and the applicable time period and processing content are determined by the start time, end time and parameter value. S12. Filter station information for the target route in the scene index record, form a station sequence according to the sort number, match the departure point and arrival point in the interval information for adjacent station names in the station sequence, write the matched interval information into the scene basic record in the order of the station sequence, and then associate the timetable information corresponding to the target route with the timetable purpose. The station information refers to the station records in the target line. The station records include the station name, network type and sequence number. The network type is used to distinguish between stations, depots, turnaround lines and storage lines. The sequence number is used to indicate the station's sequential position in the line. The sequence number refers to the positive integer position value written sequentially from the starting point to the ending point according to the direction of the target route. The current station and the next station in the station sequence are determined by the station names corresponding to the adjacent sequence numbers. Section information refers to the operation record between two adjacent stations on the target line. The section information determines the direction of the section by the departure point and arrival point, and records the section name, section length and section type; the section type is used to distinguish between mainline sections and entry / exit section sections. The departure point refers to the name of the station where the train is before entering the section in the section information, and the arrival point refers to the name of the station where the train arrives after passing through the section. When both are consistent with the names of adjacent stations in the station sequence, the section information is determined to belong to the scene basic record. S13. For each parameter configuration record corresponding to the scene identifier, match the site name with the site sequence in the scene basic record, write the corresponding sort number after matching, and determine the record writing order according to the site name, direction and start time; if the end time is earlier than the start time, write the next day mark, and form the scene configuration record according to the writing order. The direction field in the parameter configuration record defines the applicable running direction of the parameter item. When multiple parameter configuration records correspond to the same site name and the same direction, they are written in the order of start time. If the end time is earlier than the start time, it means that the parameter item crosses zero time. The next day is marked as yes, and the rest are marked as no. The scenario configuration record is the scenario parameter result formed by aggregating all parameter configuration records in this writing order.

[0006] In a preferred embodiment, S2 includes: S21. For each train in the timetable, read the arrival time, departure time, and corresponding next-day marker of each station according to the station sequence number; convert the arrival time and departure time into seconds from 0:00; when the next-day marker corresponding to a certain arrival time or departure time is yes, add 86,400 to the converted seconds to obtain the absolute arrival time and absolute departure time of each station; according to the sequence number of adjacent stations within the train, match the absolute departure time of the previous station with the absolute arrival time of the next station for the interval information between them; if no interval information is matched or the absolute departure time of the previous station is later than the absolute arrival time of the next station, write the corresponding train into the timetable error record. The timetable refers to a collection of train operation records that are linked to the target line and the purpose of the timetable. Each train record includes the train number, stops, arrival time, departure time, direction of travel, previous train number, and next train number, which are used to reflect the order of train operation on the target line. Absolute time refers to the cumulative number of seconds starting from 00:00 on the current operating day; the hours, minutes and seconds of arrival time or departure time are converted into seconds and added together. When marked as "yes" on the next day, 86,400 are added to ensure that trains crossing 00:00 are on the same time axis as trains on the current day. S22. Match the network type in the station information with the first and last stations of the train number respectively. Write the departure mark when the first station is a depot or parking lot, and write the return mark when the last station is a depot or parking lot. Write the peak-to-standby mark when the last station is not a depot or parking lot and the next train number is empty. Write the standby-peak mark when the first station is not a depot or parking lot and the previous train number is empty. In the train numbers with the same station name and running direction, sort them in ascending order by the absolute arrival time. Subtract the absolute arrival time of the current train from the absolute arrival time of the next train and write the difference into the train interval of the corresponding station of the current train. Among them, the depot refers to the line network point used for train departure, return, parking and preparation. When the first station of a train number belongs to the depot, it means that the train has entered the main line from the depot. When the last station of a train number belongs to the depot, it means that the train has returned to the depot from the main line. The parking lot is treated as a network point for trains to enter and exit the main line in the same way. The train number is an identifier that distinguishes a train operation in the timetable. The same train number corresponds to an operation record arranged in order of stations. The previous train number and the next train number are used to indicate the preceding and following sequence of the same train operation, respectively.

[0007] In a preferred embodiment, S2 further includes: S23. Read the departure point, departure point, rest point and shift change point from the scene configuration record to form scene boundary stations. If the termination position of the current train or the starting position of the next train does not belong to the scene boundary station and the two trains have the same service number, merge them sequentially along the next train number until the termination position or the starting position belongs to the scene boundary station. When the network type in the station information is a turnaround line, replace the train termination position with the station before the turnaround line. If the departure mark or return mark exists, write the train corresponding to the current train number and the next train number into the same merged train group. The departure point is the station where crew members begin their route handover task, the departure point is the station where crew members end their route handover task, the rest point is the station where crew members get off and rest between trains, and the shift change point is the station used to delineate the handover positions between adjacent shifts; when any station in the scene configuration record is written for the above purposes, that station is written as a scene boundary station. A service number is an identifier used to identify the same train entity's operational chain. Adjacent train numbers with the same service number indicate that the train takes over the next train without vehicle replacement after the previous train terminates. The same service number is used to determine the merging relationship of original turnaround trains. A turnaround line is a line network point where a train changes its direction of travel after arriving and enters the reverse train route. When the turnaround line is not used as the termination point of the crew route, the station before the turnaround line is used as the termination point of the train route, so that the termination point can correspond with the station and section information of the main line. S24. Connect the train numbers in the merged train number group in absolute time order, take the first station and the last station of the connection result and write them into the start position and end position respectively, and write the connection result, train interval and running status mark into the train number to be scheduled record. The operation status flag refers to the status result written into the train schedule record based on the first station, last station, previous train number, and next train number. It is used to indicate whether the train is in the departure, return, peak-to-standby, or standby-to-peak state, and is read by subsequent trains when taking over and generating routes.

[0008] In a preferred embodiment, S3 includes: S31. Read the train schedule records with the first absolute time order and the train schedule records with the second absolute time order one by one. When the end position is the same as the start position, write the end time of the previous train schedule as the passage completion time. When the end position is different from the start position, if the scenario configuration record contains the corresponding passage relationship, filter the passing trains that pass through the end position and the start position from the route timetable. Sort the records of passing trains whose departure time at the end position is not earlier than the end time and whose arrival time at the start position is not later than the start time by the absolute arrival time. Take the first absolute arrival time and write it as the passage completion time. Output the train schedule record. S32. Calculate the connection interval based on the train number and the opposite of the start time of the next train to be scheduled, using the end time of the previous train to be scheduled minus the start time of the next train to be scheduled. Calculate the rest time by subtracting the completion time of passage from the start time of the next train to be scheduled. If the completion time of passage is not later than the start time of the next train to be scheduled, write the previous train to be scheduled, the next train to be scheduled, the connection interval, and the rest time into the candidate connection edge, and then aggregate the candidate connection edges according to the previous train to be scheduled to form a candidate connection graph.

[0009] In a preferred embodiment, S4 includes: S41. Write the unselected state for each candidate connection edge in the candidate connection graph. During the route construction process, read the selected state of the preceding and following train numbers of the candidate connection edge. When the preceding train number has been accepted by the following train number or the following train number has been accepted by the preceding train number, keep the candidate connection edge in the unselected state. When neither the preceding nor the following train number has formed a connection relationship, write the candidate connection edge in the pending selection state and output the constrained candidate connection graph. The route construction process refers to taking the candidate connecting edge in the pending state as the receiving object, and simultaneously writing the subsequent receiving position of the previous train to be scheduled and the previous receiving position of the subsequent train to be scheduled into the already received state; taking the train to be scheduled without a previous receiving position as the starting train of the route, reading the subsequent trains to be scheduled in sequence along the already received state, until the train to be scheduled without a subsequent receiving position ends, thus forming a route; The selected state refers to the writing state of the connection relationship of the train to be scheduled, including the previous connection position state and the subsequent connection position state. When the candidate connection edge does not participate in the route construction, the previous connection position state and the subsequent connection position state are both empty. When the candidate connection edge is written to the already accepted state, the previous train to be scheduled is written to the subsequent connection position, and the subsequent train to be scheduled is written to the previous connection position. S42. Execute the resource constraint label correction algorithm using the candidate continuation graph. Start forming route labels from the candidate continuation edges in the candidate state to be selected. After each connection, accumulate the running time of the subsequent train to be scheduled to the current value multiplied by the running time, and write the rest time in the candidate continuation edge as the current rest time. If the current value multiplied by the running time exceeds the value multiplied by the running time of the corresponding shift in the scenario configuration record, or if the current rest time is less than the rest time of the corresponding shift in the scenario configuration record, delete the route label. When two route labels terminate at the same train to be scheduled, first sort them in ascending order by the accumulated value multiplied by the running time, then sort them in descending order by the current rest time if the accumulated value multiplied by the running time is the same, then sort them in ascending order by the accumulated connection interval if the accumulated connection interval is the same, and then sort them by the character order of the train number string in the route label. Keep the first route label in the sorting and output the set of feasible route paths. The shift type refers to the time interval of the route, which is divided by the morning and afternoon shift change times and the afternoon and evening shift change times recorded in the scene configuration. When the starting time of the route's starting train falls within the time interval before the morning and afternoon shift change times, it is written as the morning shift; when it falls within the time interval between the morning and afternoon shift change times and the afternoon and evening shift change times, it is written as the afternoon shift; when it falls within the time interval after the afternoon and evening shift change times, it is written as the evening shift.

[0010] In a preferred embodiment, S4 further includes: S43. Write an initial pheromone value of 1 for each candidate continuation edge in the candidate continuation graph. Multiply the pheromone values ​​of each candidate continuation edge in each feasible route in the feasible route set to obtain the path pheromone value. Add the continuation intervals of each candidate continuation edge in the path, add one, and take the reciprocal to write the path heuristic value. For each feasible route containing unscheduled trains, calculate the proportion of the product of the path pheromone value and the path heuristic value to the sum of the corresponding products of each path and write it as the path selection probability. Select feasible routes according to the path selection probability and write the unscheduled trains in the path as scheduled trains until there are no unscheduled trains. Output the route candidate scheme. Unscheduled trains refer to trains in the train schedule record that have not yet been written into any route candidate scheme; once a train enters a selected feasible route, it is written as a scheduled train and removed from the calculation of subsequent route selection probabilities. S44. First, count the number of routes for candidate routes. Then, calculate the sum of the absolute values ​​of the differences between the value multiplied by the duration of the same type of route and the average value multiplied by the duration of the same type of route. Next, calculate the sum of the absolute values ​​of the differences between the rest duration of each candidate connecting edge and the average rest duration of the candidate routes. Arrange the candidate routes in the order of the three calculation results. Backtrack the sorted candidate routes one by one. If the number of times the train to be scheduled appears is not once or the candidate connecting edge is not written into the candidate connecting graph, delete the candidate route. Add pheromones to the candidate connecting edges contained in the first-ranked candidate route according to the reverse of the ranking position. When the candidate connecting edge sets of the first-ranked candidate routes in two adjacent rounds of sorting are the same, output the initial route ranking record.

[0011] In a preferred embodiment, S5 includes: S51. Read the initial route arrangement record, arrange the routes according to the absolute start time of the first train of each route within the shift, and write the arrangement position into the route mark; when the route contains the peak-to-standby mark or standby-to-peak mark, write the standby mark; when the route contains the meal replacement entry position and the meal replacement exit position, write the train between the two into the meal replacement mark, and output the route mark record. The "substitute meal entry position" refers to the first pick-up position in the substitute meal train sequence, which is determined by the entry station name, direction of travel, absolute entry time, and entry train number. After the original train completes the pick-up at the entry position, the substitute meal train continues the service. The "substitute meal exit position" refers to the last pick-up position in the same substitute meal train sequence, which is determined by the exit station name, direction of travel, absolute exit time, and exit train number. After the substitute meal train completes the pick-up at the exit position, the subsequent train continues the service. S52. For each route in the route marking record, if the starting position of the first train is not equal to the departure point, read the trains that travel from the departure point to the starting position of the first train from the route timetable and write them into the departure marking. If the ending position of the last train is not equal to the return point, read the trains that travel from the ending position of the last train to the return point and write them into the return marking. When the same train is referenced by two or more routes, write the number of referenced routes into the passenger marking of the train. If the ending position of the last train is not equal to the terminal station of the corresponding running direction, write it into the intermediate station handover marking and output the special train marking record.

[0012] In a preferred embodiment, S5 further includes: S53. Using the train numbers in the route timetable as the rows of the arrangement verification matrix, and the routes in the special train number mark records as the columns of the arrangement verification matrix, when writing the train numbers contained in the train number record to be arranged in the route into the route, increment the corresponding row and column position once. Read the cumulative number of the arrangement verification matrix row by row. When the cumulative number is equal to one, write the arrangement pass status. When the cumulative number is zero or not one, write the arrangement abnormal status. Output the train number arrangement verification record. S54. Based on the special train number marking record, check the candidate connection map one by one for adjacent train numbers. When the termination position, start position and connection interval are consistent with the candidate connection edge, write the connection status. Then, accumulate the running time of each train in the route to obtain the shift duration, and form a rest duration sequence from the rest duration of each candidate connection edge. Compare the shift duration and rest duration sequence with the corresponding content of the initial route record item by item. When all comparison results are written to the pass status, collect the number of routes and the total mileage by shift, and divide the total mileage by the number of routes to obtain the average mileage. Output the route table optimization analysis result.

[0013] A route table optimization and parsing system for different business operation models, comprising a scenario configuration module, a time-based parsing module, a continuation mapping module, a route arrangement module, and a result verification module: The scenario configuration module is used to read the target route, timetable purpose and operation mode, retrieve the scenario identifier, route timetable and rule parameters, and associate and write them into the scenario configuration record; The timetable parsing module is used to read the route timetable one by one according to the scenario configuration record, convert the arrival time or departure time into absolute time according to the next day mark, merge adjacent trains according to the train number continuation relationship and splicing rules, and write them into the train to be scheduled record. The continuation mapping module is used to match the termination position and termination time of the preceding train to be scheduled and the starting position and starting time of the following train to be scheduled according to absolute time, calculate the connection interval and rest duration according to the traffic relationship, and write the candidate connection edges to form a candidate continuation map. The route scheduling module is used to limit the preceding and following positions of the train to be scheduled to one. It calculates the value of the candidate connecting edge multiplied by the duration, rest duration, and departure / arrival position relationship, and deletes candidate connecting edges that violate the scenario configuration record. The subsequent train to be scheduled is used as the ant colony algorithm node. The transition probability is determined by the product of pheromone quantity and the inverse of the connecting interval. If there are no unused candidate connecting edges, a route is established for the train to be scheduled that has never been scheduled. The pheromone is updated by sorting the number of routes and the discrete results of the duration. When the first candidate connecting edge set of two adjacent rounds is consistent, the initial route scheduling record is output. The result verification module is used to write the shift and special train number markings based on the initial route scheduling record, check the route timetable to determine the number of train number arrangements, compare the termination position, start position and connection interval of adjacent trains, multiply the absolute time verification value of the route by the duration and rest duration, and output the route table optimization and parsing result.

[0014] The technical effects and advantages of this invention are as follows: 1. This solution binds the operation mode to the scenario configuration record and resolves special train schedules as pending train schedule records, so that the train schedule continuation calculation has a unified basis for time, location and parameters, reducing the situation of repeated continuation, missed continuation and inconsistent rest information when switching operation parameters; 2. Convert the arrival and departure times to absolute times, and then combine them with the service number, turnaround position, and departure / return mark to merge train numbers, so that cross-zero time, turnaround, and entry / exit section operation relationships are included in the same train number record, reducing the workload of manually splitting special train numbers; 3. Calculate the connection interval and rest time based on the termination position, starting position and traffic relationship, and save the train number pair relationship in the candidate connection map so that the same station connection and the train travel connection adopt the same data structure, which is convenient for unified reading in subsequent route arrangement. 4. By screening candidate route schemes through the occupancy of receiving positions, correction of resource constraint labels and updating of pheromones, the preceding and following receiving positions of each train to be scheduled are kept to be singular, and the shift duty duration, rest duration and departure-arrival relationship are incorporated into the route construction process. 5. Verify the number of times the passenger ride is written using the arrangement and verification matrix, then check the candidate connecting edges and the rest time of the passenger ride to separate the passenger ride departure and departure times from the passenger ride times, and output a route table with complete shift, special train number markings and statistical results. Attached Figure Description

[0015] Figure 1 This is a flowchart of the route table optimization and parsing process of the present invention; Figure 2 This is a diagram of the route table optimization and analysis system of the present invention. Detailed Implementation

[0016] 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.

[0017] Refer to the instruction manual appendix Figure 1-2 The routing table optimization and parsing method of the present invention for different business operation models includes: S1. Read the target route, timetable purpose and operation mode, retrieve the scene identifier, route timetable and rule parameters, and associate and write them into the scene configuration record; Scene identifiers are used to write the basic data and operating parameters of the same target route under different operating modes into independent scene ranges. The processing equipment associates the scene basic record and scene configuration record with the scene identifier, so that train parsing, candidate connecting edge calculation, route arrangement and route verification can read data within the same scene range. The scenario configuration record includes scenario identifier, parameter type, station name, sort number, direction, start time, end time, next day marker, parameter value, and record writing order. The scenario basic record includes timetable information identifier, station sequence, and interval information. The processing equipment reads the line name from the line information, the timetable purpose from the timetable information, and the shift division, handover location, departure and arrival methods, and train operation organization relationship from the operation organization configuration. Based on this, it determines the target line, timetable purpose, and operation mode. The target line, timetable purpose, and operation mode are connected sequentially with fixed separators to form a scene identifier, and the scene index record is queried using the scene identifier. During the creation of scene index records, S11 directly outputs the scene index records that have been retrieved; if no scene index records are retrieved, the processing device writes the target line, timetable purpose and operation mode to form a scene index record. When a scene creation request carries a source scene identifier, the processing device reads the parameter configuration record corresponding to the source scene identifier, copies the parameter type, site name, direction, start time, end time, next day marker, parameter value and source write order, and then replaces the source scene identifier with the scene identifier. When the source scenario identifier does not match the parameter configuration record, the scenario index record is written to the state of no source parameter, so that the scenario management interface can write the parameter configuration record. After the scene index record is written, the processing device filters the station information according to the target line, forms a forward station sequence in ascending order of sort number, and forms a reverse station sequence in descending order of sort number; the network point type in the station information is written as station, depot, turnaround line or storage line, and the parking lot is processed as storage line network point type. During the formation of the scene basic record, S12 sequentially reads the station names corresponding to adjacent sort numbers in each station sequence, matches the origin point in the interval information with the previous station name, and matches the destination point in the interval information with the next scan point name, and writes the matched interval name, interval length and interval type into the scene basic record according to the station sequence order; if no interval information is matched, an interval missing record is written. The processing equipment then associates timetable information with the target route and timetable purpose, and writes the timetable information identifier, forward station sequence, reverse station sequence and interval information into the scenario basic record; if no timetable information is associated, it writes the timetable missing record, and the scenario index record is not included in the train reading process. After the parameter configuration record enters the positioning process, the processing equipment matches the station names in the parameters of attendance point, departure point, handover location, shift change point, rest point, meal replacement point, depot section pairing, last bus stay, and same-direction boarding to the main line to the corresponding station sequence in the running direction, and writes the sequence number into the parameter configuration record; the departure method and peak period configuration are written into the departure method or peak period with parameter values, and the depot section pairing is written into the first network point with the station name and into the second network point with the parameter value; After completing the site location, S13 converts the start and end times into the current day's time value. If the end time is earlier than the start time, it is marked as "yes" and written to the next day. If the end time is equal to or later than the start time, it is marked as "no" and written to the next day. Then, records are written in ascending order by site name, direction, and start time. If the three fields are the same, they are arranged according to the source. When the site name does not match the site sequence, the direction does not match the running direction, or the parameter value is missing, the processing device writes a parameter positioning missing record, but does not write the corresponding parameter configuration record into the scene configuration record. After the scenario configuration is recorded, the splicing rules are used for train number merging and reading, the passage relationship is used for candidate connecting edge calculation and reading, the shift and duty rest configuration are used for route arrangement and reading, and the attendance point, departure point and meal replacement point are used for route verification and reading. After processing, the bidirectional station relationships, timetable information and operation parameters of the target line are kept under the same scenario identifier, and the cross-zero time parameters, depot segment pairing and same-direction boarding mainline parameters have a definite reading position; For example, the processing equipment establishes a scene identifier for the weekday operation mode of the target line, copies the attendance point, departure point, rest point, shift change point, meal replacement point and depot section matching parameters in the existing weekday scene, compares the end time of the night stay parameter with the start time and writes it to the next day marked as yes, and then provides the formed scene configuration record to cross-zero hour train parsing and route connection calculation.

[0018] S2. Read the route timetable for each train according to the scenario configuration record, convert the arrival time or departure time to the absolute time according to the next day mark, merge adjacent trains according to the train number connection relationship and splicing rules, and write them into the train to be scheduled record. After the route timetable is associated with the scene identifier, target route and timetable purpose, the processing equipment converts the original train number records into train number records to be scheduled with a unified time scale, operating status and train number merging relationship; Each train record in the route timetable is determined by the train number, direction of travel, and station order. The station records respectively save the arrival time, departure time, and the corresponding next-day marker of the route timetable. The next-day marker of the route timetable is only used for train time conversion and is stored separately from the next-day marker in the scenario configuration record. When the timetable is parsed, the processing equipment reads the arrival time, departure time and the next day mark of the timetable according to the train number and station order. It converts the hours, minutes and seconds into seconds and adds them together. The next day mark of the timetable is hour plus 86,400. The absolute arrival time and the absolute departure time are written respectively. In the time conversion process, S21 uses the departure absolute time of the originating station as the originating station time and the arrival absolute time of the destination station as the destination station time. When the arrival absolute time of an intermediate station is later than the departure absolute time, or the departure absolute time of the previous station is later than the arrival absolute time of the next station, the processing equipment writes the train number, station name, previous absolute time, next absolute time, and reverse order type into the timetable error record. The processing equipment then reads the names of adjacent stations within the train according to the direction of travel, and matches the departure and arrival points in the basic scenario record with the names of adjacent stations. If no interval information is matched, it writes a timetable error record. Trains in the timetable error record are not written into the subsequent train status record. Trains that have been matched are written into a timetable parsing record containing the absolute arrival time, absolute departure time, and interval information. After the absolute time is written into the timetable parsing record, the processing equipment obtains the first and last stations of each train according to the station order, and identifies the depot or parking lot by the depot or storage line network type in the station information; if the first station belongs to the depot or parking lot, a departure mark is written, and if the last station belongs to the depot or parking lot, a return mark is written. During the train status writing process, S22 writes a peak-to-standby flag when the train's final station does not belong to the depot or parking lot and the next train number is empty, and writes a standby-to-peak flag when the train's first station does not belong to the depot or parking lot and the previous train number is empty. The departure flag, return flag, peak-to-standby flag, and standby-to-peak flag are written as parallel fields of the operation status flags in the train status record. The processing equipment then aggregates train status records by station name and direction of travel, sorts them in ascending order by absolute arrival time, and sorts them by train number when the absolute arrival times are the same. The difference between the absolute arrival time of the next train and the absolute arrival time of the current train is written into the headway of the station corresponding to the current train. The headway of the last item in the train group is written as a null value. The attendance points, departure points, rest points, and shift change points in the scene configuration record are filtered by station name and direction and then written to the scene boundary station. The scene boundary station is used to cut off the merging range of the original turnaround trains. S23 selects the train number that has not been written into the merged train number group from the timetable parsing record as the current train number, and reads the end position of the current train number, the start position of the train number corresponding to the next train number, and the service numbers of the two train numbers. If neither the current train's termination position nor the next train's start position has been written to the scene boundary station, the service number is the same, and the next train number has been matched with a train record, then the next train will be written to the current merged train group, and the written next train will be used as the new current train to continue reading. When the next train number is empty, the service number is different, or the end position or start position is written to the scene boundary station, the processing device ends the current merged train group; when the next train number points to a train that has been written into the current merged train group, the train number is written in a loop record and the current merged train group ends. When the train's termination position is on a turnaround line, the processing equipment reads the station name preceding the turnaround line's sequence number in the current direction of travel and replaces the train's termination position; when the turnaround line is at the beginning of the station sequence, a record of missing turnaround positions is written. When an exit mark or return mark exists and the next train number matches the train number record, the processing equipment writes the current train number and the corresponding train number of the next train number into the same merged train number group, and retains the original turnaround merge relationship or entry / exit merge relationship in the merged train number group; After the combined train number group is written, the processing equipment concatenates the train numbers in ascending order of their start absolute time. If the arrival absolute times are the same, they are arranged by train number. The first station in the concatenation result is written to the start position, the last station is written to the end position, the start absolute time of the first train is written to the start absolute time, and the end absolute time of the last train is written to the end absolute time. S24 calculates the running time by subtracting the starting absolute time from the ending absolute time, and writes the concatenation result, headway, running status marker, starting position, ending position, starting absolute time, ending absolute time, and running time into the train schedule record to be scheduled; when the merged train group contains only one train, the first station, last station, and absolute time of the train itself are written into the train schedule record to be scheduled. Trains associated with timetable anomalies, train number loops, or missing turnaround locations are not written into the pending train record, but are re-entered into the parsing process after the timetable is corrected. The train schedule record is formed by the original train time, station interval, running status and merging relationship. The subsequent candidate continuation edge calculation can directly read the termination position, start position, termination absolute time, start absolute time and running time. For example, if a train departs from the previous station at 23:50 and arrives at the next station at 00:10 the following day, the processing equipment will write the departure time as 85,800 seconds and add 86,400 to the arrival time to write it as 87,000 seconds. If the service number of a train is the same as the service number of the next train and the turnaround position is not written to the scene boundary station, the two trains will be written into the same merged train group in absolute time order to form a new train record to be scheduled.

[0019] S3. Match the termination position and termination time of the preceding train to be scheduled and the start position and start time of the following train to be scheduled according to the absolute time. Calculate the connection interval and rest duration according to the traffic relationship and write them into the candidate connection edge to form a candidate connection graph. The connection between the pending train records is determined by the end position, the start position and the corresponding absolute time. The processing equipment first forms a train pair passage record, and then writes the train pairs with the passage completion time into the candidate connection diagram. The passage relationship in the scene configuration record saves the passage start position, passage end position, permitted passage direction and passage purpose; after the same direction of travel on the main line parameter is written into the passage relationship, it is used to limit the range of train numbers that the crew can read when traveling from the start point, end point or route end point to another location; The train schedule records are arranged in ascending order of their starting absolute time. If the starting absolute times are the same, they are arranged according to the train schedule record identifier. The processing device sequentially selects the train schedule record with the earlier order as the preceding train schedule record and the train schedule record with the later order as the following train schedule record. It also reads the end position and end absolute time of the preceding train schedule record and the start position and start absolute time of the following train schedule record. During the process of writing the train number to the passage record, when the termination position is the same as the start position, S31 writes the absolute termination time of the previous train number to be scheduled into the passage completion time; when the termination position is different from the start position, the processing device matches the passage relationship in the scenario configuration record with the termination position and the start position, and then filters the passing trains that pass through the termination position and the start position in sequence from the route timetable. If the absolute departure time of a passing train at its termination position is earlier than the absolute termination time of the preceding train to be scheduled, or if the absolute arrival time of a passing train at its starting position is later than the absolute start time of the following train to be scheduled, it will not be written into the train pair's passage record; the remaining passing trains will be arranged in ascending order of their absolute arrival times, and if the absolute arrival times are the same, they will be arranged in order of their passing train number, and the passing train number, departure absolute time, arrival absolute time, and passage completion time corresponding to the arrival absolute time will be written into the train pair's passage record. When the scene configuration record does not match the passage relationship or no passing train is filtered, the processing device writes an unpassable record, and the preceding and subsequent trains to be scheduled do not enter the candidate continuation edge writing process. After the train number is recorded, the processing equipment calculates the connection interval by subtracting the absolute time of the end of the previous train from the absolute time of the start of the next train to be scheduled, and calculates the rest time by subtracting the time of completion of passage from the absolute time of the start of the next train to be scheduled. When the completion time of passage is earlier than or equal to the absolute start time of the subsequent train to be scheduled, write the identifier of the preceding train to be scheduled, the identifier of the subsequent train to be scheduled, the previous end position, the subsequent start position, the connection interval, the rest duration, the passing train number, the absolute departure time of passage, and the absolute arrival time of passage into the candidate connection edge. During the writing of candidate continuation edges, S32 uses the previous and subsequent train number identifiers as candidate continuation edge identifiers. Only one candidate continuation edge is retained for the same candidate continuation edge identifier, and candidate continuation edges are collected according to the previous train number identifier. The train number record to be scheduled is used as the vertex of the candidate continuation graph, and the candidate continuation edge is used as a directed edge from the previous train number to the subsequent train number. Unpassable records associated with pending train numbers are still retained as candidate continuation graph vertices, for use as start or end train numbers during route scheduling; passable train numbers are only written to candidate continuation edges, not to the value of the pending train number record or the train number string. The candidate connection graph enables the same location connection and the same vehicle passage connection to calculate the rest time using the same passage completion time field, and subsequent route arrangement can directly read the train connection relationship according to the directed edge. For example, if the preceding train to be scheduled ends at station A at 8:00, and the following train to be scheduled starts at station B at 8:30, and there is a train in the timetable that departs from station A at 8:05 and arrives at station B at 8:20, the processing equipment will write 8:20 as the completion time, 30 minutes as the connection interval, 10 minutes as the rest period, and write the candidate connection edge from the preceding train to the following train.

[0020] S4. Limit the number of trains to be scheduled to one position before and after each train. Calculate the value of the candidate connecting edge by multiplying the duration, rest duration, and departure / arrival position relationship. Delete candidate connecting edges that violate the scenario configuration record. Use the subsequent trains to be scheduled as ant colony algorithm nodes. Determine the transition probability by multiplying the pheromone quantity by the inverse of the connecting interval. If there are no unused candidate connecting edges, establish a route for the un-arranged trains to be scheduled. Update the pheromone by sorting the number of routes and the discrete results of the duration. Output the initial route arrangement record when the first candidate connecting edge sets of two adjacent rounds are consistent. After the candidate connection graph is written to the candidate connection edge, the processing device controls the vehicle occupancy relationship based on the preceding and following connection positions of the vehicle to be scheduled, and forms the initial route scheduling record within the scope of the shift, duty duration, rest duration, attendance position and departure position in the same scenario configuration record. The vertices in the candidate continuation graph are constrained to retain the identifier of the train number to be scheduled, and the candidate continuation edges retain the identifiers of the preceding and following train numbers to be scheduled, the continuation interval, the rest duration, and the information of the passing trains. At the start of each round of route construction, the processing equipment writes each candidate connection edge in the candidate connection diagram into an unselected state and clears the previous and subsequent connection position status of each train to be scheduled; then it reads the previous and subsequent train to be scheduled identifiers of the candidate connection edges one by one. When the preceding train number of the candidate continuation edge has been written into the subsequent receiving position, or when the subsequent train number of the candidate continuation edge has been written into the preceding receiving position, the candidate continuation edge remains unselected; when both receiving positions are empty, the processing device writes the candidate continuation edge into the selectable state. When a candidate connecting edge is selected for a route, S41 simultaneously writes the subsequent receiving position of the preceding train to be scheduled as the subsequent train to be scheduled identifier, writes the preceding receiving position of the subsequent train to be scheduled as the preceding train to be scheduled identifier, and then writes the candidate connecting edge into the already received state; when any receiving position has been written into another train to be scheduled identifier, the writing of the candidate connecting edge this time is cancelled and the unselected state is maintained. The processing equipment starts with the train number record to be scheduled when the previous receiving position is empty, and reads the train number record to be scheduled along the subsequent receiving position. When the subsequent receiving position is empty, the route reading ends, and the train number string and the candidate connecting edge identifier string are written into the route construction record. If a train number record to be scheduled is read again during the reading process, it is written into the route loop record, and the corresponding candidate connecting edge is not written into the selection state. The route label is formed by the candidate connecting edges in the pending selection state and the pending train number records that have not been written to the preceding receiving position; The processing equipment writes the date (early shift, middle shift, or night shift) based on the relationship between the absolute start time of the route's starting train and the shift change times (morning / midnight or afternoon / evening). Then, it reads the shift's duty duration, rest duration, start point, end point, and departure / arrival method from the scenario configuration record. If the starting position of the route's starting train is consistent with the start point, or if there is a travel relationship from the start point to the starting position in the scenario configuration record, the attendance relationship is written. During the resource constraint label correction process, S42 establishes a route label based on the starting train number. The route label includes the starting train number identifier, the current terminating train number identifier, the train number string, the shift, the cumulative shift duration, the current rest duration, the cumulative connection interval, and the previous route label identifier. The initial value of the cumulative shift duration is written with the running time of the starting train number, the current rest duration is written as a null value, and the cumulative connection interval is written as zero. Each time a candidate connecting edge in the receiving and selection state is received, the processing equipment adds the running time of the subsequent train to the cumulative value multiplied by the running time, writes the rest time of the candidate connecting edge into the current rest time, adds the connecting interval of the candidate connecting edge to the cumulative connecting interval, and writes the connecting identifier of the subsequent train to the train number string; if the cumulative value multiplied by the running time is greater than the running time of the corresponding shift, or if the current rest time is less than the rest time of the corresponding shift, the route tag is deleted. When a candidate connecting edge cannot continue to accept, the processing device matches the termination position of the current terminating train to the return point or the passage relationship from the termination position to the return point. If no return relationship is matched, the route label is deleted. When two route labels terminate at the same train to be scheduled, they are sorted in ascending order by cumulative value multiplied by duration, descending order by current rest duration, ascending order by cumulative connecting interval, and in the order of train number string characters. The first route label in the sorting is retained and the set of feasible route paths is output. After pheromone initialization, the processing device writes an initial pheromone value of 1 to each candidate continuation edge in the candidate continuation graph, and multiplies the pheromone amounts of each candidate continuation edge in the feasible route path to form the path pheromone amount; the continuation intervals of each candidate continuation edge in the path are added together and then one second is added, and the reciprocal is taken to form the path heuristic value. For feasible routes for a single train that do not contain candidate continuation edges, the path pheromone amount 1 and the path heuristic value 1 are written. When constructing route candidate schemes, S43 establishes the same number of ant construction sequence numbers as the number of trains to be scheduled. At the beginning of the ant construction sequence number, the scheduled state of unscheduled trains to be scheduled is cleared. The processing device calculates the product of the path pheromone quantity and the path heuristic value from the feasible route paths containing unscheduled trains to be scheduled, and then divides the product of each path by the sum of the products of all paths and writes it into the path selection probability. Feasible routes are arranged in the order of train number strings and the route selection probability is accumulated sequentially. The pseudo-random number generator forms a seed with the round number and ant construction number and outputs a value between zero and one. The processing device selects the feasible route whose accumulated route selection probability first reaches the pseudo-random number and writes the train numbers to be scheduled in the route into the already arranged train numbers. If there is no feasible route that includes unscheduled trains, the next route is established using the first character of the train number string in the unscheduled trains. After all train schedules to be scheduled are written into the scheduled status, the processing equipment writes the routes selected by the same ant construction number into the route candidate scheme and clears the scheduled status to proceed to the next ant construction number. After the candidate route schemes are completed, the processing equipment first counts the number of routes, then collects the cumulative value of routes by shift, multiplies the duration by the cumulative value of each shift, divides the sum of the cumulative values ​​multiplied by the duration of each shift by the number of routes in the same shift to form the average value multiplied by the duration of each shift, and accumulates the absolute value of the difference between the cumulative value multiplied by the duration of each route in the same shift and the average value multiplied by the duration of each shift. The processing equipment only collects the rest time for the selected candidate connecting edges in the route candidate scheme. It divides the sum of the rest times by the number of candidate connecting edges to form the average rest time. Then, it sums the absolute value of the difference between the rest time of each candidate connecting edge and the average rest time. When no candidate connecting edges are selected in the route candidate scheme, the sum of the absolute values ​​of the rest time differences is written as zero. S44 sorts the candidate route schemes in ascending order by the number of routes, in ascending order by the sum of the absolute values ​​of the time difference of the same shift, and in ascending order by the sum of the absolute values ​​of the rest time difference. If the three calculation results are the same, they are sorted according to the character order of the train number string in the candidate route scheme. The train number records to be scheduled in the candidate route scheme are read one by one. If the number of train number records to be scheduled is zero or more than one, or if the selected candidate connecting edge does not match the candidate connecting graph, the candidate route scheme is deleted. The processing device increments the pheromone content of each candidate connecting edge in the first-ranked route candidate scheme by one, and forms a candidate connecting edge set with the previous and subsequent train number identifiers of the candidate connecting edge; when the members and number of members of the candidate connecting edge set of the first-ranked route candidate schemes in two adjacent rounds are the same, the first-ranked route candidate scheme is written into the route initial scheduling record. Through the occupancy of receiving positions, correction of resource constraint labels, selection of path probabilities, and back lookup of route candidate schemes, each train record to be scheduled in the initial route scheduling record has a preceding or following relationship, and maintains the relationship of shift, duty duration, rest duration, and departure / arrival positions. For example, after the running time of a certain train to be scheduled is written into the early shift duty time, another train to be scheduled is taken over by the candidate connecting edge. The processing equipment adds the running time of the subsequent trains to the cumulative duty time. When the rest time of the candidate connecting edge meets the early shift rest time, the route label is retained. When two route labels terminate at the same train to be scheduled, one route label is retained according to the four-item order. Then, the initial route scheduling record is formed by the consistent set of candidate connecting edges of the adjacent two rounds of route candidate schemes. S5. Based on the initial route schedule record, write the shift type and special train number mark, check the route timetable to determine the number of train number arrangements, compare the termination position, start position and connection interval of adjacent trains, multiply the absolute time check value of the route by the duration and rest duration, and output the route table optimization parsing result. The processing equipment reads the initial route scheduling record. The route in the initial route scheduling record includes the identifier string of the train number to be scheduled, shift, candidate connecting edge identifier string, cumulative shift duration and rest duration sequence. The train number to be scheduled record retains the original train number of the merged train group. The processing equipment collects routes by shift, sorts them in ascending order by the absolute start time of the first train and the train number of the first train, and writes the sorting order into the route marker. When writing the route marker, S51 reads the peak-to-standby marker and the standby-to-peak marker. If either marker exists, it writes the standby marker. Then, it matches the meal replacement point parameters in the scenario configuration record with the meal replacement entry position and meal replacement exit position. If the train number following the substitute entry position and the train number preceding the substitute exit position are on the same route, and the entry position's take-up order is earlier than the exit position's take-up order, the train number records between the two are written into the substitute entry mark; otherwise, they are written into the substitute entry position missing record; when the last train's standby parameter matches the last train of the route, a standby mark is written, forming a route mark record. The processing equipment matches the shift, direction of travel, and absolute start time of the first train or absolute end time of the last train in the route marking record with the scenario configuration record to determine the attendance location and departure location. When the starting position of the first train is inconsistent with the departure point, the trains that pass through the departure point and the starting position of the first train and whose absolute arrival time is no later than the absolute start time of the first train are selected from the route timetable. They are sorted in ascending order by absolute arrival time and train number, and the first item of the sort is written into the travel attendance mark. When the last train's termination location is inconsistent with the departure point, filter the trains that depart from the last train's termination location and arrive at the departure point, sort them in ascending order by absolute departure time and train number, and write the first item in the sorting to the departure mark. If no trains are found, write the missing attendance record or missing departure record. When the same train number is associated with two or more routes, S52 writes the associated route identifier string and the number of associated routes into the passenger-carrying mark; when the last train's termination position is inconsistent with the last station in the running direction station sequence, it writes the intermediate station handover mark; the train attendance mark, train departure mark, passenger-carrying mark and intermediate station handover mark together form a special train number mark record; S53 uses the route timetable information identifier and the original train number to form the row identifier of the arrangement verification matrix, and the route identifier to form the column identifier. The position of each row and column is written as zero. The original train number contained in the train number to be arranged in the route is read one by one. The position of the original train number corresponding to the row and the corresponding column of the route identifier is incremented by one. The train number associated with the passenger attendance mark, passenger departure mark and passenger-carrying mark is not written into the arrangement verification matrix. The processing device accumulates the column position values ​​row by row. When the cumulative number equals one, it writes the arrangement pass status. When the cumulative number is zero, it writes the unarranged status. When the cumulative number is greater than one, it writes the duplicate arrangement status, thus forming a train number arrangement verification record. The processing equipment retrieves candidate connection diagrams based on the preceding and following train IDs of adjacent train ID records. It then compares the termination position, start position, and connection interval of the candidate connection edge with the route marking records item by item. If the three fields match, it writes "connection passed" status; otherwise, it writes "connection failed" status. S54 accumulates the running time of the trains to be scheduled within the route according to the order of acceptance to form a value-multiplied duration. It writes the rest time between adjacent trains to be scheduled according to the same order to form a rest time sequence, and compares it item by item with the cumulative value-multiplied duration and rest time sequence in the initial route scheduling record; the rest time sequence of a single train route is written as an empty sequence. When both the train schedule verification record and the acceptance verification result are written to the "passed" status, the processing equipment collects the number of routes by shift, sums the lengths of the intervals traversed by the trains within the routes to form the total mileage, divides the total mileage by the number of routes to form the average mileage, and writes it into the route table optimization analysis result; when no verification result is written to the "passed" status, the route identifier, the train number to be scheduled identifier, the verification item, and the verification status are written to the route verification record, but the route is not written into the route table optimization analysis result.

[0021] exist Figure 1 It should be noted that: parallelograms represent input data, rule parameters, records formed during processing, or final results; rounded rectangles represent modules that perform processing; solid arrows indicate that data is transmitted in the direction of the arrow; the scene configuration module, timetable analysis module, continuous mapping module, route arrangement module, and result verification module sequentially complete scene configuration, timetable analysis, train connection, route arrangement, and result verification; dashed boxes are used to collect parameters such as shift change, rest, on-duty, and shift duration read by route arrangement, and do not indicate the transmission direction. exist Figure 2 It should be noted that: the outer rounded corner box represents the system scope, the left rectangle represents the functional module, the upper long rectangle represents the display of the route table optimization and parsing results, and the upward arrow indicates the output of the results; the middle double-headed arrow box represents the record reading and result writing relationship between modules, the dashed rectangle represents the data recording range of scene configuration records, train schedule records, candidate connection diagrams and initial route scheduling records, and the bottom solid rectangle represents the data management range of route timetables, rule parameters and scene configuration records.

[0022] Furthermore, the present invention also includes a route table optimization and parsing processing system for different business operation modes, the system comprising a scenario configuration module, a time-based parsing module, a continuous mapping module, a route arrangement module, and a result verification module. The scenario configuration module is used to read the target route, timetable purpose and operation mode, retrieve the scenario identifier, route timetable and rule parameters, and associate and write them into the scenario configuration record; The timetable parsing module is used to read the route timetable one by one according to the scenario configuration record, convert the arrival time or departure time into absolute time according to the next day mark, merge adjacent trains according to the train number continuation relationship and splicing rules, and write them into the train to be scheduled record. The continuation mapping module is used to match the termination position and termination time of the preceding train to be scheduled and the starting position and starting time of the following train to be scheduled according to absolute time, calculate the connection interval and rest duration according to the traffic relationship, and write the candidate connection edges to form a candidate continuation map. The route scheduling module is used to limit the preceding and following positions of the train to be scheduled to one. It calculates the value of the candidate connecting edge multiplied by the duration, rest duration, and departure / arrival position relationship, and deletes candidate connecting edges that violate the scenario configuration record. The subsequent train to be scheduled is used as the ant colony algorithm node. The transition probability is determined by the product of pheromone quantity and the inverse of the connecting interval. If there are no unused candidate connecting edges, a route is established for the train to be scheduled that has never been scheduled. The pheromone is updated by sorting the number of routes and the discrete results of the duration. When the first candidate connecting edge set of two adjacent rounds is consistent, the initial route scheduling record is output. The result verification module is used to write the shift and special train number markings based on the initial route scheduling record, check the route timetable to determine the number of train number arrangements, compare the termination position, start position and connection interval of adjacent trains, multiply the absolute time verification value of the route by the duration and rest duration, and output the route table optimization and parsing result.

[0023] Overall execution and implementation instructions: The system first establishes scenario configuration records based on the target route, timetable purpose, and operation mode, binding parameters such as stations, sections, departure and arrival, shift changes, rest, and meal replacements to the same scenario; then, it converts the arrival and departure times in the route timetable into absolute times, identifies return trips, peak-to-standby trips, standby-to-peak trips, and turnaround trips, and merges them into pending trip records according to service number and scenario boundaries; The processing equipment then calculates the connection interval and rest time based on the termination position, start position and traffic relationship of the train to be scheduled, forming a candidate connection map. In the route scheduling stage, the connection relationship between the trains to be scheduled is first restricted by the receiving position, and then the paths that do not conform to the scenario configuration are filtered out by the shift duration, rest time and departure and return relationship. The path selection probability is calculated by pheromone and connection interval, and route candidate schemes are formed round by round. Finally, the number of train schedules, the connection between adjacent trains, the duty duration and the rest duration are checked back, and the route table optimization analysis results with class and special train number markings are output. For example, if a subway line has trains that depart at midnight, trains that leave the depot, and trains that are switched to standby during peak hours on weekday nights, the system first writes the trains that leave at 11:50 p.m. and the trains that arrive at 12:10 a.m. the next day into the same absolute time axis, and then merges the turnaround trains with the same service number that have not passed through the shift change point. When the previous scheduled train ends at station A and the next scheduled train begins at station B, the system filters trains from station A to station B from the timetable and calculates the rest time for crew members after arriving at station B; trains that meet the requirements of early shift duty time and rest time are included in the route candidate scheme. After the route is completed, the system will write the record of the passenger taking the train from the departure point to the starting position of the first train into the passenger attendance mark, and confirm that each passenger taking the train will only be recorded as one route, thus forming a route table that can be directly used for crew organization.

[0024] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing and parsing routing tables for different business operation models, characterized in that, include: S1. Read the target route, timetable purpose and operation mode, retrieve the scene identifier, route timetable and rule parameters, and associate and write them into the scene configuration record; S2. Read the route timetable for each train according to the scenario configuration record, convert the arrival time or departure time to the absolute time according to the next day mark, merge adjacent trains according to the train number connection relationship and splicing rules, and write them into the train to be scheduled record. S3. Match the termination position and termination time of the preceding train to be scheduled and the start position and start time of the following train to be scheduled according to the absolute time. Calculate the connection interval and rest duration according to the traffic relationship and write them into the candidate connection edge to form a candidate connection graph. S4. Limit the number of trains to be scheduled to one position before and after each train. Calculate the value of the candidate connecting edge by multiplying the duration, rest duration, and departure / arrival position relationship. Delete candidate connecting edges that violate the scenario configuration record. Use the subsequent trains to be scheduled as ant colony algorithm nodes. Determine the transition probability by multiplying the pheromone quantity by the inverse of the connecting interval. If there are no unused candidate connecting edges, establish a route for the un-arranged trains to be scheduled. Update the pheromone by sorting the number of routes and the discrete results of the duration. Output the initial route arrangement record when the first candidate connecting edge sets of two adjacent rounds are consistent. S5. Based on the initial route schedule record, write the shift type and special train number mark, check the route timetable to determine the number of train number arrangements, compare the termination position, start position and connection interval of adjacent trains, multiply the absolute time check value of the route by the duration and rest duration, and output the route table optimization analysis result.

2. The routing table optimization and parsing method for different business operation models according to claim 1, characterized in that: S1 includes: S11. The target route, timetable purpose and operation mode are written in a fixed order to form a scenario identifier. When the source scenario identifier exists, the corresponding parameter configuration record is read and copied. The source scenario identifier in the copied record is replaced with the scenario identifier, and the scenario index record is output. S12. Filter station information for the target route in the scene index record, form a station sequence according to the sort number, match the departure point and arrival point in the interval information for adjacent station names in the station sequence, write the matched interval information into the scene basic record in the order of the station sequence, and then associate the timetable information corresponding to the target route with the timetable purpose. S13. For each parameter configuration record corresponding to the scene identifier, match the site name with the site sequence in the scene basic record, write the corresponding sort number after matching, and determine the record writing order according to the site name, direction and start time; if the end time is earlier than the start time, write the next day mark, and form the scene configuration record according to the writing order.

3. The routing table optimization and parsing method for different business operation models according to claim 2, characterized in that: S2 includes: S21. For each train in the timetable, read the arrival time, departure time, and corresponding next-day marker of each station according to the station sequence number; convert the arrival time and departure time into seconds from 0:00; when the next-day marker corresponding to a certain arrival time or departure time is yes, add 86,400 to the converted seconds to obtain the absolute arrival time and absolute departure time of each station; according to the sequence number of adjacent stations within the train, match the absolute departure time of the previous station with the absolute arrival time of the next station for the interval information between them; if no interval information is matched or the absolute departure time of the previous station is later than the absolute arrival time of the next station, write the corresponding train into the timetable error record. S22. Match the network type in the station information with the first and last stations of the train number. Write the departure mark when the first station is a depot or parking lot, and write the return mark when the last station is a depot or parking lot. Write the peak-to-standby mark when the last station is not a depot or parking lot and the next train number is empty. Write the standby-peak mark when the first station is not a depot or parking lot and the previous train number is empty. In the train numbers with the same station name and running direction, sort them in ascending order by the absolute arrival time. Subtract the absolute arrival time of the current train from the absolute arrival time of the next train and write the difference into the headway of the corresponding station of the current train number.

4. The routing table optimization and parsing method for different business operation models according to claim 3, characterized in that: S2 also includes: S23. Read the departure point, departure point, rest point and shift change point from the scene configuration record to form scene boundary stations. If the termination position of the current train or the starting position of the next train does not belong to the scene boundary station and the two trains have the same service number, merge them sequentially along the next train number until the termination position or the starting position belongs to the scene boundary station. When the network type in the station information is a turnaround line, replace the train termination position with the station before the turnaround line. If the departure mark or return mark exists, write the train corresponding to the current train number and the next train number into the same merged train group. S24. Connect the train numbers in the merged train number group in absolute time order, take the first station and the last station of the connection result and write them into the start position and end position respectively, and write the connection result, train interval and running status mark into the train number to be scheduled record.

5. The routing table optimization and parsing method for different business operation models according to claim 4, characterized in that: S3 includes: S31. Read the train schedule records with the first absolute time order and the train schedule records with the second absolute time order one by one. When the end position is the same as the start position, write the end time of the previous train schedule as the passage completion time. When the end position is different from the start position, if the scenario configuration record contains the corresponding passage relationship, filter the passing trains that pass through the end position and the start position from the route timetable. Sort the records of passing trains whose departure time at the end position is not earlier than the end time and whose arrival time at the start position is not later than the start time by the absolute arrival time. Take the first absolute arrival time and write it as the passage completion time. Output the train schedule record. S32. Calculate the connection interval based on the train number and the opposite of the start time of the next train to be scheduled, using the end time of the previous train to be scheduled minus the start time of the next train to be scheduled. Calculate the rest time by subtracting the completion time of passage from the start time of the next train to be scheduled. If the completion time of passage is not later than the start time of the next train to be scheduled, write the previous train to be scheduled, the next train to be scheduled, the connection interval, and the rest time into the candidate connection edge, and then aggregate the candidate connection edges according to the previous train to be scheduled to form a candidate connection graph.

6. The routing table optimization and parsing method for different business operation models according to claim 5, characterized in that: S4 includes: S41. Write the unselected state for each candidate connection edge in the candidate connection graph. During the route construction process, read the selected state of the preceding and following train numbers of the candidate connection edge. When the preceding train number has been accepted by the following train number or the following train number has been accepted by the preceding train number, keep the candidate connection edge in the unselected state. When neither the preceding nor the following train number has formed a connection relationship, write the candidate connection edge in the pending selection state and output the constrained candidate connection graph. S42. Execute the resource constraint label correction algorithm using the constrained candidate connection graph. Start forming route labels from the candidate connection edges in the candidate state to be selected. After each connection, accumulate the running time of the subsequent train to be scheduled to the current value multiplied by the running time, and write the rest time in the candidate connection edge as the current rest time. If the current value multiplied by the running time exceeds the value multiplied by the running time of the corresponding shift in the scenario configuration record, or if the current rest time is less than the rest time of the corresponding shift in the scenario configuration record, delete the route label. When two route labels terminate at the same train to be scheduled, first sort them in ascending order by the accumulated value multiplied by the running time. If the accumulated value multiplied by the running time is the same, sort them in descending order by the current rest time. If the current rest time is the same, sort them in ascending order by the accumulated connection interval. If the accumulated connection interval is the same, sort them by the character order of the train number string in the route label. Keep the first route label in the sorting and output the set of feasible route paths.

7. The routing table optimization and parsing method for different business operation models according to claim 6, characterized in that: S4 also includes: S43. Write an initial pheromone value of 1 for each candidate continuation edge in the candidate continuation graph. Multiply the pheromone values ​​of each candidate continuation edge in each feasible route in the feasible route set to obtain the path pheromone value. Add the continuation intervals of each candidate continuation edge in the path, add one, and take the reciprocal to write the path heuristic value. For each feasible route containing unscheduled trains, calculate the proportion of the product of the path pheromone value and the path heuristic value to the sum of the corresponding products of each path and write it as the path selection probability. Select feasible routes according to the path selection probability and write the unscheduled trains in the path as scheduled trains until there are no unscheduled trains. Output the route candidate scheme. S44. First, count the number of routes for candidate routes. Then, calculate the sum of the absolute values ​​of the differences between the value multiplied by the duration of the same type of route and the average value multiplied by the duration of the same type of route. Next, calculate the sum of the absolute values ​​of the differences between the rest duration of each candidate connecting edge and the average rest duration of the candidate routes. Arrange the candidate routes in the order of the three calculation results. Backtrack the sorted candidate routes one by one. If the number of times the train to be scheduled appears is not once or the candidate connecting edge is not written into the candidate connecting graph, delete the candidate route. Add pheromones to the candidate connecting edges contained in the first-ranked candidate route according to the reverse of the ranking position. When the candidate connecting edge sets of the first-ranked candidate routes in two adjacent rounds of sorting are the same, output the initial route ranking record.

8. The routing table optimization and parsing method for different business operation models according to claim 7, characterized in that: S5 includes: S51. Read the initial route arrangement record, arrange the routes according to the absolute start time of the first train of each route within the shift, and write the arrangement position into the route mark; when the route contains the peak-to-standby mark or standby-to-peak mark, write the standby mark; when the route contains the meal replacement entry position and the meal replacement exit position, write the train between the two into the meal replacement mark, and output the route mark record. S52. For each route in the route marking record, if the starting position of the first train is not equal to the departure point, read the trains that travel from the departure point to the starting position of the first train from the route timetable and write them into the departure marking. If the ending position of the last train is not equal to the return point, read the trains that travel from the ending position of the last train to the return point and write them into the return marking. When the same train is referenced by two or more routes, write the number of referenced routes into the passenger marking of the train. If the ending position of the last train is not equal to the terminal station of the corresponding running direction, write it into the intermediate station handover marking and output the special train marking record.

9. The routing table optimization and parsing method for different business operation models according to claim 8, characterized in that: S5 also includes: S53. Using the train numbers in the route timetable as the rows of the arrangement verification matrix, and the routes in the special train number mark records as the columns of the arrangement verification matrix, when writing the train numbers contained in the train number record to be arranged in the route into the route, increment the corresponding row and column position once. Read the cumulative number of the arrangement verification matrix row by row. When the cumulative number is equal to one, write the arrangement pass status. When the cumulative number is zero or not one, write the arrangement abnormal status. Output the train number arrangement verification record. S54. Based on the special train number marking record, check the candidate connection map one by one for adjacent train numbers. When the termination position, start position and connection interval are consistent with the candidate connection edge, write the connection status. Then, accumulate the running time of each train in the route to obtain the shift duration, and form a rest duration sequence from the rest duration of each candidate connection edge. Compare the shift duration and rest duration sequence with the corresponding content of the initial route record item by item. When all comparison results are written to the pass status, collect the number of routes and the total mileage by shift, and divide the total mileage by the number of routes to obtain the average mileage. Output the route table optimization analysis result.

10. A route table optimization and parsing system for different business operation modes, used to implement the route table optimization and parsing method for different business operation modes as described in any one of claims 1-9, the system comprising a scenario configuration module, a time-based parsing module, a continuous mapping module, a route arrangement module, and a result verification module, characterized in that: The scenario configuration module is used to read the target route, timetable purpose and operation mode, retrieve the scenario identifier, route timetable and rule parameters, and associate and write them into the scenario configuration record; The timetable parsing module is used to read the route timetable one by one according to the scenario configuration record, convert the arrival time or departure time into absolute time according to the next day mark, merge adjacent trains according to the train number continuation relationship and splicing rules, and write them into the train to be scheduled record. The continuation mapping module is used to match the termination position and termination time of the preceding train to be scheduled and the starting position and starting time of the following train to be scheduled according to absolute time, calculate the connection interval and rest duration according to the traffic relationship, and write the candidate connection edges to form a candidate continuation map. The route scheduling module is used to limit the preceding and following positions of the train to be scheduled to one. It calculates the value of the candidate connecting edge multiplied by the duration, rest duration, and departure / arrival position relationship, and deletes candidate connecting edges that violate the scenario configuration record. The subsequent train to be scheduled is used as the ant colony algorithm node. The transition probability is determined by the product of pheromone quantity and the inverse of the connecting interval. If there are no unused candidate connecting edges, a route is established for the train to be scheduled that has never been scheduled. The pheromone is updated by sorting the number of routes and the discrete results of the duration. When the first candidate connecting edge set of two adjacent rounds is consistent, the initial route scheduling record is output. The result verification module is used to write the shift and special train number markings based on the initial route scheduling record, check the route timetable to determine the number of train number arrangements, compare the termination position, start position and connection interval of adjacent trains, multiply the absolute time verification value of the route by the duration and rest duration, and output the route table optimization and parsing result.

Images