A truck reservation and scheduling method based on train arrival and departure time windows

Abstract

This invention discloses a truck reservation and scheduling method based on train arrival and departure time windows. The method constructs a single-path set S based on the current station, sets the preset complete time window of this path as T, and the actual complete time window of this path as T′. It obtains the complete time window difference |D| on this path, the preset time window threshold |Y| for this path, and the corresponding original band field. It then constructs gene chains with their own time series-based structures, and uses |Y| to match |Y... i The system makes judgments and adjusts the scheduling plan to achieve overall scheduling. By monitoring the trains along the entire route, the matching time between the trains and trucks is calculated and adjusted, thereby reducing the time and space pressure on the current station and making the overall matching and scheduling difficulty lower. When non-punctual operation occurs, the scheduling plan is revised by adjusting the truck reservation time in advance, which reduces scheduling costs and improves the station's operating efficiency.

Description

Technical Field

[0001] This invention relates to the field of train scheduling technology, specifically a method for pre-booking and scheduling of trucks based on train arrival and departure time windows. Background Technology

[0002] Freight trains are a common mode of land-based bulk cargo transportation. After the trains arrive, the bulk cargo needs to be transferred and distributed by trucks. However, since the trains that stop at a station are not the only ones, and the trucks used for transfer cannot wait in the station for a long time and the operating space within the station is limited, effective scheduling is required to ensure that the trucks and trains can coordinate their schedules effectively, thereby ensuring that the dwell time of the cargo and trucks at the station is within a manageable range.

[0003] The existing scheduling scheme involves multiple parties making time reservations, with vehicles waiting outside the station until the train arrives. Vehicles then enter the station around the same time as the scheduled train, and finally, cargo is transferred. However, this method is highly susceptible to resource waste and disruption of operations due to delays by any party. Because the current station hosts multiple scheduled trains, the high density and pressure of these time slots significantly increases the pressure on the station's reservation and scheduling system.

[0004] Therefore, it is necessary to design a truck reservation and scheduling method based on the arrival and departure time window of the train. Summary of the Invention

[0005] The purpose of this invention is to provide a truck reservation and scheduling method based on the arrival and departure time window of the train, so as to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A truck reservation and scheduling method based on train arrival and departure time windows includes the following steps:

[0008] S100. Construct a single path set S based on the current station, which is the set of several sub-segments of the interval from the originating station to this station. Each sub-segment consists of two adjacent stations and is denoted as s1, s2, ..., s2. i ;

[0009] S200. The preset full time window for constructing this route is T, and the preset time for each sub-segment from the originating station to this station is set as t1, t2, ..., t. i That is, T = t1 + t2 + ... + t i ;

[0010] S300. Let the actual complete time window of the path be T', and the actual time used for each sub-segment be t′1, t′2, ..., t′ i, that is, T′=t′1+t′2+…+t′ i ;

[0011] S400. By comparing the preset full time window T of the path with the actual full time window T′, the difference |D| of the full time window on the path is obtained.

[0012] S500, perform double-value recovery on the time window difference |D|, and then calculate the preset time window threshold |Y| for the path based on the time point through double-value calculation, as well as the corresponding original band domain;

[0013] S600: Gene points are constructed through all trains at this station, independent chains are constructed through all paths at this station, and then gene points are matched with chains to form gene chains with their own time series.

[0014] Based on S100 to S500, each gene point has an independent time window threshold during runtime, which, based on the time series, are |Y1|, |Y2|, ..., |Y i |, through |Y| and the close-range and distant-range domains formed by it and |Y i The decision is made, and the scheduling plan is adjusted accordingly.

[0015] S800: The number of chains that the current site can accommodate at any given time is allocated in a coordinated manner based on the time domain of gene point expansion, thereby achieving overall scheduling.

[0016] According to the above technical solution, the sub-segments of S100 are divided based on the stops of the trains. That is, on the same route, different trains stop at different adjacent stations, and the sub-segments they are divided into are different.

[0017] According to the above technical solution, the time t used for the preset sub-segment of S200 i It is the departure time interval between two adjacent stations, which includes the time from departure from the previous station to arrival at the current station, and the time from arrival at the current station to departure from the current station.

[0018] According to the above technical solution, the actual time t′ used by the sub-segment of S300 i The change in the value represents the time period from departure from the previous station to arrival at the current station, and it is affected by the current road conditions.

[0019] According to the above technical solution, the calculation of S200 and S300 through S400 shows that the difference in the complete time window |D| is composed of the sub-differences of several sub-segments, that is, |D|=|d1+d2+…+d i |, at the same time d i =(t i -t′ i), and obtain the time variation amplitude value of a single sub-segment |d i |, for |d i Perform data statistics and construct a distribution model based on |d| i |Extend the distribution, and at the same time, associate the influencing factors with the corresponding |d i We perform matching and association to form a sub-road segment database.

[0020] According to the above technical solution, the calculation of the time window difference |D| versus the time window threshold |Y| by S500 is as follows:

[0021] By using binary recovery, two values, D and -D, can be obtained from |D|. Simultaneously, calculations are performed at predetermined time points to obtain a Y value of |2D|. Thus, the range of |Y| can be determined to be (0~|2D|). Based on the range of |Y| and the predetermined time points, the original band region is formed. Since the range of |Y| is known, the band region is further divided. Let the original band region of (0~|D|) on both sides of the time point be the close band region, and let the original band region of (|D|~|2D|) on both sides of the time point be the distant band region.

[0022] According to the above technical solution, the specific operation of S600 is as follows:

[0023] Each train journey is designated as a single data point, thus each data point has a pre-set complete time window T. n The preset time t for each sub-section n The actual complete time window T′ n The actual time t′ used for the sub-section n , Complete time window difference |D n |With time window threshold|Y n |, n=1,2,…,i;

[0024] The current station has several paths, and each path will have several trains running in chronological order. That is, any path can arrange the gene points in chronological order to form a gene chain based on time sequence, with each train corresponding to a gene point.

[0025] According to the above technical solution, the specific judgment operation of S700 is as follows:

[0026] Each gene point has its own preset time window threshold |Y|. When a gene point is run again, repeating steps S100 to S500 will generate a new time window threshold |Y|. n | and the corresponding new band;

[0027] Compare |Y| with the original band generated at the expected time point and |Y n|Compare the new band with the one generated at the expected time point, that is, determine the position of the extreme value of the new band in the original band. When it is located in the close band, it is the timely operation of the gene point; when it is located in the distant band, it is the non-timely operation of the gene point.

[0028] For gene points that are not running on time, the arrival time needs to be recalculated. That is, the arrival time is preset by using the median of the new band, and the calculation steps of S100 to S500 are repeated. They are marked as floating gene points, and the corresponding original gene points are hidden to form a pseudo-on-time running state.

[0029] According to the above technical solution, the scheduling method of S800 is as follows:

[0030] S801. Based on gene point-based on on-time and pseudo-on-time operation, the preset arrival time points are coordinated and a time distribution model is formed.

[0031] S802. Assign the original banding domain to each gene point, and extend each banding domain based on the time distribution model.

[0032] S803. Based on the number of simultaneous occupancy at the current station and all paths traversed by the current station, integrate and combine them to form a combined set;

[0033] S804. The extension results of the band based on the time distribution model obtained by S802 are used to separate the overlapping gene points, and at the same time, the corresponding number of clusters are formed based on the number of sites that can be accommodated at the same time.

[0034] S805. Trains within the aforementioned cluster need to be combined and matched from the combination set to achieve effective scheduling of train stops.

[0035] S806. Based on the results of train stop allocation, the transshipment trucks are scheduled as a whole. The algorithm used is the Kalman filter algorithm, which calculates both the time value T of the train and the time value T of the transshipment truck. b With the time value T of the truck j ;

[0036] S807, By analyzing the time value T of the train b With the time value T of the truck j The comparison, i.e., T b / T j = N, when N<0.85, the train is late; when 0.85≤N≤1.15, the train and truck times match; when 1.15≤N, the truck time is late.

[0037] S808. Effective scheduling of trucks is achieved by analyzing the time matching between trucks and trains.

[0038] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0039] By monitoring the trains along the entire route and calculating and adjusting the matching time between the trains and trucks, the time and space pressure at the current station is reduced, thus lowering the overall matching and scheduling difficulty. When non-punctual operation occurs, the computational pressure at the current station is reduced by adjusting the scheduling plan, thereby reducing scheduling costs and improving the station's operational efficiency. Attached Figure Description

[0040] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0041] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, the present invention provides a technical solution: a method for scheduling container trucks based on the arrival and departure time windows of trains, comprising the following steps:

[0043] S100. Construct a single path set S based on the current station, which is the set of several sub-segments of the interval from the originating station to this station. Each sub-segment consists of two adjacent stations and is denoted as s1, s2, ..., s2. i ;

[0044] S200. The preset full time window for constructing this route is T, and the preset time for each sub-segment from the originating station to this station is set as t1, t2, ..., t. i That is, T = t1 + t2 + ... + t i ;

[0045] S300. Let the actual complete time window of the path be T', and the actual time used for each sub-segment be t′1, t′2, ..., t′ i , that is, T′=t′1+t′2+…+t′ i ;

[0046] S400. By comparing the preset full time window T of the path with the actual full time window T′, the difference |D| of the full time window on the path is obtained.

[0047] S500, perform double-value recovery on the time window difference |D|, and then calculate the preset time window threshold |Y| for the path based on the time point through double-value calculation, as well as the corresponding original band domain;

[0048] S600: Gene points are constructed through all trains at this station, independent chains are constructed through all paths at this station, and then gene points are matched with chains to form gene chains with their own time series.

[0049] Based on S100 to S500, each gene point has an independent time window threshold during runtime, which, based on the time series, are |Y1|, |Y2|, ..., |Y i |, through |Y| and the close-range and distant-range domains formed by it and |Y i The decision is made, and the scheduling plan is adjusted accordingly.

[0050] S800: The number of chains that the current site can accommodate at any given time is allocated in a coordinated manner based on the time domain of gene point expansion, thereby achieving overall scheduling.

[0051] Specifically, the sub-segments of S100 are divided based on the stops of the trains. That is, on the same route, different trains stop at different adjacent stations, and the sub-segments they are divided into are different.

[0052] Specifically, the time t used for the preset sub-segment in S200 i It is the departure time interval between two adjacent stations, which includes the time from departure from the previous station to arrival at the current station, and the time from arrival at the current station to departure from the current station.

[0053] Specifically, the actual time t′ used by the sub-segment of S300 i The change in the value represents the time period from departure from the previous station to arrival at the current station, and it is affected by the current road conditions.

[0054] Specifically, as can be seen from the calculations of S200 and S300 in S400, the difference in the complete time window |D| is composed of the sub-differences of several sub-segments, that is, |D|=|d1+d2+…+d i |, at the same time d i =(t i -t′ i ), and obtain the time variation amplitude value of a single sub-segment |d i |, for |d i Perform data statistics and construct a distribution model based on |d| i |Extend the distribution, and at the same time, associate the influencing factors with the corresponding |d i We perform matching and association to form a sub-road segment database.

[0055] Specifically, the calculation of the time window difference |D| versus the time window threshold |Y| in S500 is as follows:

[0056] By using binary recovery, two values, D and -D, can be obtained from |D|. Simultaneously, calculations are performed at predetermined time points to obtain a Y value of |2D|. Thus, the range of |Y| can be determined to be (0~|2D|). Based on the range of |Y| and the predetermined time points, the original band region is formed. Since the range of |Y| is known, the band region is further divided. Let the original band region of (0~|D|) on both sides of the time point be the close band region, and let the original band region of (|D|~|2D|) on both sides of the time point be the distant band region.

[0057] Specifically, the operation of S600 is as follows:

[0058] Each train journey is designated as a single data point, thus each data point has a pre-set complete time window T. n The preset time t for each sub-section n The actual complete time window T′ n The actual time t′ used for the sub-section n , Complete time window difference |D n |With time window threshold|Y n |, n=1,2,…,i;

[0059] The current station has several paths, and each path will have several trains running in chronological order. That is, any path can arrange the gene points in chronological order to form a gene chain based on time sequence, with each train corresponding to a gene point.

[0060] Specifically, the discrimination operation of S700 is as follows:

[0061] Each gene point has its own preset time window threshold |Y|. When a gene point is run again, repeating steps S100 to S500 will generate a new time window threshold |Y|. n | and the corresponding new band;

[0062] Compare |Y| with the original band generated at the expected time point and |Y n |Compare the new band with the one generated at the expected time point, that is, determine the position of the extreme value of the new band in the original band. When it is located in the close band, it is the timely operation of the gene point; when it is located in the distant band, it is the non-timely operation of the gene point.

[0063] For gene points that are not running on time, the arrival time needs to be recalculated. That is, the arrival time is preset by using the median of the new band, and the calculation steps of S100 to S500 are repeated. They are marked as floating gene points, and the corresponding original gene points are hidden to form a pseudo-on-time running state.

[0064] Specifically, the scheduling method of S800 is as follows:

[0065] S801. Based on gene point-based on on-time and pseudo-on-time operation, the preset arrival time points are coordinated and a time distribution model is formed;

[0066] S802. Assign the original banding domain to each gene point, and extend each banding domain based on the time distribution model.

[0067] S803. Based on the number of simultaneous occupancy of the current station and all paths traversed by the current station, integrate and combine them to form a combined set;

[0068] S804. The extension results of the band based on the time distribution model obtained by S802 are used to separate the overlapping gene points, and at the same time, the corresponding number of clusters are formed based on the number of sites that can be accommodated at the same time.

[0069] S805. Trains within the aforementioned cluster need to be combined and matched from the combination set to achieve effective scheduling of train stops.

[0070] S806. Based on the results of train stop allocation, the transshipment trucks are scheduled as a whole. The algorithm used is the Kalman filter algorithm, which calculates both the time value T of the train and the time value T of the transshipment truck. b With the time value T of the truck j ;

[0071] S807, By analyzing the time value T of the train b With the time value T of the truck j The comparison, i.e., T b / T j = N, when N<0.85, the train is late; when 0.85≤N≤1.15, the train and truck times match; when 1.15≤N, the truck time is late.

[0072] S808. Effective scheduling of trucks is achieved by analyzing the time matching between trucks and trains.

[0073] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0074] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. 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 pre-booking and scheduling container trucks based on train arrival and departure time windows, characterized in that, Includes the following steps: S100. Construct a single path set S based on the current station, which is the set of several sub-segments of the interval from the originating station to this station. Each sub-segment consists of two adjacent stations and is denoted as s1, s2, ..., s i ; S200. The preset complete time window for constructing this route is T, and the preset time for each sub-segment from the originating station to this station is set as t1, t2, ..., t. i That is, T = t1 + t2 + ... + t i ; S300. Let the actual complete time window of the path be T′, and the actual time used for each sub-segment be t′1, t′2, ..., t′. i , that is, T′=t′1+t′2+...+t′ i ; S400. By comparing the preset full time window T of the path with the actual full time window T′, the difference |D| of the full time window on the path is obtained. S500, perform double-value recovery on the time window difference |D|, and then calculate based on the time point using double values ​​to obtain the preset time window threshold |Y| for the path, as well as the corresponding original band domain; S600: Gene points are constructed through all trains at this station, independent chains are constructed through all paths at this station, and then gene points are matched with chains to form gene chains with their own time series. Based on S100 to S500, each gene point has an independent time window threshold during runtime, which, based on the time series, are |Y1|, |Y2|, ..., |Y i |, through |Y| and the close-range and distant-range domains formed by it and |Y i The decision is made, and the scheduling plan is adjusted accordingly. S800: The number of chains that the current station can accommodate at the same time is allocated in a coordinated manner based on the time domain of gene point expansion, thereby realizing the coordinated scheduling of trucks and trains.

2. The truck reservation and scheduling method based on train arrival and departure time windows according to claim 1, characterized in that, The sub-segments of S100 are divided based on the stops of the trains. That is, on the same route, different trains stop at different adjacent stations, and the sub-segments they are divided into are different.

3. The truck reservation and scheduling method based on train arrival and departure time windows according to claim 1, characterized in that, The time t used for the preset sub-segment of S200 i It is the departure time interval between two adjacent stations, which includes the time from departure from the previous station to arrival at the current station, and the time from arrival at the current station to departure from the current station.

4. The truck reservation and scheduling method based on train arrival and departure time windows according to claim 1, characterized in that, The actual time t′ used by the sub-segment of S300 i The change in the value represents the time period from departure from the previous station to arrival at the current station, and it is affected by the current road conditions.

5. The truck reservation and scheduling method based on train arrival and departure time windows according to claim 1, characterized in that, The calculations of S200 and S300, as described in S400, show that the difference in the complete time window |D| is composed of the sub-differences of several sub-segments, i.e., |D|=|d1+d2+...+d i |, at the same time d i =(t i -t′ i ), and obtain the time variation amplitude value of a single sub-segment |d i |, for |d i Perform data statistics and construct a distribution model based on |d| i |Extend the distribution, and at the same time, associate the influencing factors with the corresponding |d i We perform matching and association to form a sub-road segment database.

6. The truck reservation and scheduling method based on train arrival and departure time windows according to claim 1, characterized in that, The calculation of the time window difference |D| versus the time window threshold |Y| by S500 is as follows: By using binary recovery, two values, D and -D, can be obtained from |D|. Simultaneously, calculations are performed at predetermined time points to obtain a Y value of |2D|. Thus, the range of |Y| can be determined to be (0~|2D|). Based on the range of |Y| and the predetermined time points, the original band region is formed. Since the range of |Y| is known, the band region is further divided. Let the original band region of (0~|D|) on both sides of the time point be the close band region, and let the original band region of (|D|~|2D|) on both sides of the time point be the distant band region.

7. The truck reservation and scheduling method based on train arrival and departure time windows according to claim 1, characterized in that, The specific operation of S600 is as follows: Each train journey is designated as a single data point, thus each data point has a pre-set complete time window T. n The preset time t for each sub-section n The actual complete time window T′ n The actual time t′ used for the sub-section n , Complete time window difference |D n |With time window threshold|Y n |, n = 1, 2, ..., i; The current station has several paths, and each path will have several trains running in chronological order. That is, any path can arrange the gene points in chronological order to form a gene chain based on time sequence, with each train corresponding to a gene point.

8. The truck reservation and scheduling method based on train arrival and departure time windows according to claim 7, characterized in that, The specific operation of the S700 for discrimination is as follows: Each gene point has its own preset time window threshold |Y|. When a gene point is run again, repeating steps S100 to S500 will generate a new time window threshold |Y|. n | and the corresponding new band; Compare |Y| with the original band generated at the expected time point and |Y n |Compare the new band with the one generated at the expected time point, that is, determine the position of the extreme value of the new band in the original band. When it is located in the close band, it is the timely operation of the gene point; when it is located in the distant band, it is the non-timely operation of the gene point. For gene points that are not running on time, the arrival time needs to be recalculated. That is, the arrival time is preset by using the median of the new band, and the calculation steps of S100 to S500 are repeated. They are marked as floating gene points, and the corresponding original gene points are hidden to form a pseudo-on-time running state.

9. A truck reservation and scheduling method based on train arrival and departure time windows according to claim 8, characterized in that, The scheduling method of the S800 is as follows: S801. Based on gene point-based on on-time and pseudo-on-time operation, the preset arrival time points are coordinated and a time distribution model is formed; S802. Assign the original banding domain to each gene point, and extend each banding domain based on the time distribution model. S803. Based on the number of simultaneous occupancy of the current station and all paths traversed by the current station, integrate and combine them to form a combined set; S804. The extension results of the band based on the time distribution model obtained by S802 are used to separate the overlapping gene points, and at the same time, the corresponding number of clusters are formed based on the number of sites that can be accommodated at the same time. S805. Trains within the aforementioned cluster need to be combined and matched from the combination set to achieve effective scheduling of train stops. S806. Based on the results of train stop allocation, the transshipment trucks are scheduled as a whole. The algorithm used is the Kalman filter algorithm, which calculates both the time value T of the train and the time value T of the transshipment truck. b With the time value T of the truck i ; S807, By analyzing the time value T of the train b With the time value T of the truck j The comparison, i.e., T b / T j = N, when N < 0.85, the train is late; when 0.85 ≤ N ≤ 1.15, the train and truck times match; when 1.15 ≤ N, the truck time is late. S808. Effective scheduling of trucks is achieved by analyzing the time matching between trucks and trains.

Images