Container ship rolling scheduling optimization system in sleeve mooring hot joining mode

Abstract

The optimization system provided by the invention comprises model construction and a rolling scheduling optimization algorithm, in the rolling scheduling optimization algorithm, when the number of ships is large, a heuristic algorithm is selected for solving, all the ships are grouped according to the arrival date, all the ships arriving in one day are scheduled together, and the scheduling efficiency is improved. The port entering sequence and berth distribution of the arriving ships on the second day are influenced by the scheduling result of the scheduled ships on the previous day; the following basic assumptions are made to simplify the construction process of the optimization model: all berths are discrete berths; the predicted arrival time, the workload and the tonnage of the container ships arriving at the same day are known in advance; the operation efficiency of the quay crane of each berth is fixed; ship sailing in the same direction cannot follow over. According to the method, a scheduling optimization model with the purpose of minimizing the total waiting time is established. The provided multi-round rolling scheduling algorithm can effectively solve the optimal scheduling scheme of continuously arriving ships within a period of time, and the channel utilization rate is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ship scheduling, and particularly to an optimization system for the rolling scheduling of container ships in the tandem berthing hot swap mode. Background Art

[0002] As the core hub of international logistics and supply chains, the operating efficiency of ports is directly related to the operational effectiveness of the entire logistics system. How to effectively improve berth utilization and reduce ship waiting time has become a core issue in port logistics research. In this context, ship scheduling optimization has become an important means to solve the above problems. By establishing accurate mathematical models and applying advanced algorithms, ship scheduling optimization can significantly reduce waiting time, improve transportation efficiency, and optimize the resource allocation of ports.

[0003] However, the scheduling optimization models in the prior art often ignore external limiting factors in specific environments during design. In fact, in some specific port environments, ship scheduling problems need to consider the influence of external conditions such as weather, sea conditions, and tides, which are crucial in the optimization process. "Tandem berthing hot swap" is an efficient port scheduling mode that enables efficient operation through seamless connection of ships. Although the "tandem berthing hot swap" mode has been proven to effectively alleviate port congestion and reduce waiting time, there is little research on the "tandem berthing hot swap" mode in existing related solutions, and there is a lack of joint research on ship scheduling optimization and berth allocation in the "tandem berthing hot swap" mode. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide an optimization system for the rolling scheduling of container ships in the tandem berthing hot swap mode to solve the problems raised in the above background art. The present invention establishes a scheduling optimization model with the goal of minimizing the total waiting time. The proposed multi-round rolling scheduling algorithm can effectively solve the optimal scheduling plan for continuously arriving ships within a certain period.

[0005] To achieve the above purpose, the present invention is realized through the following technical solutions: an optimization system for the rolling scheduling of container ships in the tandem berthing hot swap mode, the optimization system includes model construction and a rolling scheduling optimization algorithm. When the number of ships is large in the rolling scheduling optimization algorithm, a heuristic algorithm is selected for solution, and all ships are grouped according to the arrival date. All ships arriving within one day are scheduled together, and the inbound order and berth allocation of ships arriving the next day are affected by the scheduling results of the ships scheduled the previous day; the following basic assumptions are made during the model construction process to simplify the construction process of the optimization model: the berth types are all discrete berths; the expected arrival time, workload, and ship tonnage of container ships arriving on the same day are known in advance; the quay crane operation efficiency of each berth is fixed; ships sailing in the same direction cannot overtake each other.

[0006] Furthermore, during the model construction process, based on the objectives and constraints of the scheduling under the "nesting hot connection" mode, the formula for establishing the optimization model is as follows:

[0007]

[0008] T b -T i ≥M(1-X ib ); i ∈ N, b ∈ B (6)

[0009]

[0010]

[0011] X ib ,F ij ,D i ,E iq ∈ {0,1} (13)

[0012] Furthermore, in the formula of the optimization model, the ship set N = {1, 2,... n}; the berth set B = {1, 2,... b}; T i is the cargo capacity of ship i; U is the berth operation efficiency; t safe is the ship safety distance (expressed in time); is the arrival time of ship i; is the entry time of ship i into the port; is the departure time of ship i from the berth; t anc is the sailing time of the ship from the anchorage to the berth; [TS, TE] represents the traffic control period, during which ships are prohibited from entering the port starting from TS and the traffic control ends at TE;.

[0013] Furthermore, T b is the load limit of berth b; t tidestart is the start time of the high tide period; t interval is the interval time of each high tide period; t tideend is the end time of the high tide period; Q is the high tide period set; M is a sufficiently large positive integer; X ib takes 1 when ship i goes to berth b, otherwise takes 0; F ij takes 1 when ship j enters the port after ship i, otherwise takes 0; D i takes 1 when ship i needs to enter the port at high tide, otherwise takes 0; E iq takes 1 when ship i enters the port during the q tide period, otherwise takes 0.

[0014] Furthermore, Equation (1) is the objective function, representing the minimum total waiting time of all ships; Equation (2) indicates that the start time of entering the port should be later than the arrival time; Equation (3) shows that the departure time of a ship from the berth is the sum of its start time of entering the port, the channel navigation time, and the berth operation time; Equation (4) means that according to port regulations, a safe distance needs to be maintained between two adjacent ships entering the channel; Equation (5) is the constraint on the entering order of ships entering the port successively; Equation (6) requires that the ship tonnage should match the berth length.

[0015] Furthermore, Equation (7) states that large-tonnage ships must enter the port during the high-tide period; Equations (8) and (9) are the time constraints for ships taking advantage of the rising tide and must follow the time constraints of the high-tide period; Equation (10) is the constraint of the "berth tandem operation" mode. When the terminal dispatching and command center confirms that the remaining time of the departing ship does not exceed the time for the berthed ship to reach the berth from the anchorage, it can notify the following ship to enter the port; Equations (11)-(12) are for traffic control and channel closure due to bad weather. During the traffic control period, no ship can enter the port.

[0016] Furthermore, the process of the rolling scheduling optimization algorithm is as follows: Ship data is grouped by date; global variables are initialized; ships are processed date by date; an initial population is randomly generated, where n represents the number of convergence times, initially 0; the fitness value is evaluated and the optimal individual is recorded; it is calculated whether the difference from the best fitness value of the previous generation is less than the threshold.

[0017] Furthermore, if the difference from the best fitness value of the previous generation is less than the threshold, then convergence is checked through n = n + 1. It is determined whether n is greater than the convergence lower limit. If n is greater than the convergence lower limit, the optimal ship scheduling result for the current day is output, feasibility verification is carried out, and the global variables are updated. The scheduling plan for the current day is accumulated, and finally the optimal scheduling plan for all ships is output.

[0018] Furthermore, if n is less than the convergence lower limit, then through the elite selection strategy, two-point crossover strategy, and crossover mutation strategy in sequence, a new population is generated, and the process of evaluating the fitness value and recording the optimal individual is returned.

[0019] Furthermore, if the difference from the best fitness value of the previous generation is greater than the threshold, then n = 0, and convergence is checked, and it continues to be determined whether n is greater than the convergence lower limit.

[0020] The beneficial effects of the present invention:

[0021] 1. The container ship rolling scheduling optimization system under the berth tandem operation mode comprehensively considers the scheduling of the entering order of container ships and berth allocation, combines factors such as "berth tandem operation", channel driving rules, tidal restrictions, and bad weather, and establishes a scheduling optimization model with the goal of minimizing the total waiting time. The proposed multi-round rolling scheduling algorithm can effectively solve the optimal scheduling plan for ships arriving continuously within a period of time.

[0022] 2. The rolling scheduling optimization system for container ships under the tandem hot-swap mode significantly reduces the average waiting time of ships, improves the port service capacity, provides an effective solution for optimizing port scheduling, increases the utilization rate of waterways, and alleviates the congestion in the anchorage area, which is of great significance for improving port logistics efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flowchart of the rolling scheduling optimization algorithm of the present invention;

[0024] Figure 2 It is a graph showing the changing trend of the average waiting time of the FCFS model and the rolling scheduling model in the embodiment of the present invention;

[0025] Figure 3 It is a distribution diagram of the waiting time of ships under the FCFS and rolling scheduling models in the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0026] In order to make the technical means, creative features, achieved objectives and functions of the present invention easy to understand, the present invention will be further described below in conjunction with specific embodiments.

[0027] Please refer to Figures 1 to 3 , the present invention provides the following technical solutions: a rolling scheduling optimization system for container ships under the tandem hot-swap mode. In this embodiment, in order to simplify the construction process of the optimization model, the following basic assumptions are made in this study:

[0028] 1) All berth types are discrete berths;

[0029] 2) The expected arrival time, workload and ship tonnage of the container ships arriving on the same day are known in advance;

[0030] 3) The quay crane operation efficiency of each berth is fixed;

[0031] 4) Ships sailing in the same direction cannot overtake.

[0032] And based on the above assumptions, and based on the objectives and constraints of the scheduling under the "tandem hot-swap" mode, the following optimization model is established:

[0033]

[0034] T b -T i ≥M(1-X ib ); i∈N,b∈B (6)

[0035]

[0036]

[0037] Xib , F ij , D i , E iq ∈ {0, 1} (13)

[0038] The variables and symbols in the above formula provided by this embodiment are defined as follows:

[0039] The set of ships N = {1, 2, … n}; the set of berths B = {1, 2, … b}; T i is the cargo capacity of ship i; U is the berth operation efficiency; t safe is the safety distance between ships (expressed in time); is the arrival time of ship i at the port; is the time when ship i enters the port; is the time when ship i leaves the berth; t anc is the navigation time of the ship from the anchorage to the berth; [TS, TE] represents the traffic control period, during which ships are prohibited from entering the port starting from TS and the traffic control ends at TE; T b is the load limit of berth b; t tidestart is the start time of the high tide period; t interval is the interval time of each high tide period; t tideend is the end time of the high tide period; Q is the set of high tide periods; M is a sufficiently large positive integer. X ib takes 1 when ship i goes to berth b, otherwise takes 0; F ij takes 1 when ship j enters the port after ship i, otherwise takes 0; D i takes 1 when ship i needs to enter the port at high tide, otherwise takes 0; E iq takes 1 when ship i enters the port during the q tide period, otherwise takes 0.

[0040] Equation (1) is the objective function, indicating the minimum total waiting time of all ships; Equation (2) means that the start time of entering the port should be later than the arrival time; Equation (3) means that the time when the ship leaves the berth is the sum of its start time of entering the port, the navigation time in the channel and the berth operation time; Equation (4) means that according to the port regulations, a safety distance needs to be maintained between two adjacent ships entering the channel; Equation (5) is the constraint on the order of entering the port of ships entering the port successively; Equation (6) means that the ship tonnage should match the berth length; Equation (7) means that large-tonnage ships must enter the port during the high tide period; Equations (8) and (9) are the time constraints for ships entering the port at high tide; Equation (10) is the "berth hot connection" mode constraint. When the terminal dispatching and command center confirms that the remaining time of the departing ship does not exceed the time for the berthing ship to reach the berth from the anchorage, it can notify the following ship to enter the port. Equations (11)-(12) are for traffic control and channel closure due to bad weather. During the traffic control period, no ship can enter the port.

[0041] This embodiment also provides a rolling scheduling optimization algorithm, such asFigure 1 As shown in Figure 1 , in the "berth tandem operation" mode, the connection between ships becomes closer. The arrival time at the port and the departure time of the leading ship, as well as the navigation time in the channel and the arrival time at the berth of the following ship, all need to be perfectly coordinated and scheduled to maximize the berth utilization rate. Therefore, only studying the scheduling of ships arriving within a short period of time cannot reflect the optimization effect of the "berth tandem operation" mode. How to optimize the scheduling of continuously arriving ships within a month, so the impact of the scheduling results of the previous day needs to be considered in the scheduling of each day. Therefore, a heuristic algorithm for rolling ship scheduling optimization is proposed to solve this optimization model. In the rolling scheduling optimization algorithm, all ships are grouped according to the arrival date. All ships arriving within a day are scheduled together, and the arrival order and berth allocation of ships arriving the next day will be affected by the scheduling results of the ships scheduled the previous day, and so on.

[0042] This embodiment also provides the container terminal of Qinzhou Port in Guangxi Zhuang Autonomous Region as an experimental scenario, and designs an experiment to compare the results of the First Come First Serve (FCFS) model and the proposed rolling scheduling optimization, as follows:

[0043] 1. Experimental parameter settings

[0044] According to historical data analysis, the arrival time interval of container ships at Qinzhou Port in 2022 conforms to a Poisson distribution with a mean of 1.87 hours. Based on this, a time series of ship arrivals at the anchorage is randomly generated in the simulation experiment. At this time, the port provides three types of berths for ships to choose: Berths 1-4 serve small ships, berths 5-8 serve medium-sized ships, and berths 9-10 serve large ships. The navigation time for ships from the anchorage to the berth is about 2 hours. For safety considerations, the time interval between two adjacent ships in the channel should be greater than 30 minutes. Large container ships need to enter the port at high tide. The start time of the first high-tide period is 6 o'clock and the end time is 10 o'clock, and the high-tide period interval is 8 hours. In addition, temporary traffic control is required due to reduced visibility caused by bad weather and large ships occupying the channel, suspending ship arrivals. In the experiment, several traffic control periods will be randomly generated to simulate such situations.

[0045] 2. Analysis of experimental results

[0046] Considering the scheduling problem in the case of the increasing number of arriving ships with the development of the port. First, the average waiting time of ships under different arrival rates is calculated through the rolling scheduling algorithm and compared with the FCFS model, and a comprehensive comparative analysis of the performance of the two scheduling methods is carried out.

[0047] Figure 2The changing trends of the average waiting times of two scheduling modes under different arrival rates are shown. It can be seen that as the number of ships increases, the average waiting times of both scheduling models increase. However, the rolling scheduling model is more efficient than the FCFS model in handling ship scheduling because it shows a lower average waiting time under the same arrival rate, and the service ship threshold of the rolling scheduling model is also higher than that of the FCFS model. For the FCFS model, the service ship threshold is approximately 88 min / ship. When the arrival rate is lower than this threshold, the average waiting time shows a slow linear increasing trend. When the ship arrival rate is higher than this threshold, the average waiting time increases exponentially with the increase in the number of ships. For the rolling scheduling model, the service arrival rate threshold is approximately 80 min / ship. When it is lower than this threshold, the average waiting time also shows a slow increasing trend. When it is higher than this threshold, the average waiting time also increases rapidly, but the increase amplitude is much smaller than that of the FCFS model. This also proves that the rolling scheduling model has significant advantages compared with the FCFS model under high ship arrival rates. In addition, the error bars in the figure reflect the confidence intervals of the 10 simulation results. It can be seen that the interval of the FCFS model is wider than that of the rolling scheduling model, especially when the ship arrival rate is high. This indicates that the rolling scheduling model has higher stability than the FCFS model in terms of the average waiting time.

[0048] Figure 3 The waiting time distributions of the FCFS and rolling scheduling models under different arrival rates are shown. The data results are the means of 10 simulation experiments. Figure 3 As can be seen from Fig. a, when the ship arrival rate is 88 min / ship, under both models, the waiting times of most ships are concentrated at relatively small values, while the waiting times of a few ships are distributed at relatively large values. In the FCFS model, 51.3% of the ships have a waiting time less than 100 min before entering the waterway, while in the rolling scheduling model, it is 68.1%. This indicates that the rolling scheduling model allows more ships to enter the waterway with a shorter waiting time. Figure 3 As can be seen from Fig. b, when the ship arrival rate is 80 min / ship, the waiting times of the ships under the FCFS model are fitted by kernel density estimation and show a multi-modal distribution. The waiting times are widely distributed in the range of 0 to 3000 min, specifically concentrated around 800 min, and the tail of the waiting time drops relatively slowly, which means that the waiting times of some ships are relatively long. Under the rolling scheduling mode, the waiting times of most ships are in a relatively small interval. Although there are a small number of ships with waiting times exceeding 800 min, it can still reduce the average waiting time of all ships to 245 min, avoiding long waiting times for a large number of ships. Figure 3It can be seen that when the monthly ship arrival volume increases from 489 to 545, the waiting time under the FCFS model increases sharply, from 271 minutes to 1089 minutes. While the rolling scheduling model controls the ship waiting time within a reasonable range by optimizing and adjusting the arrival order of container ships of different tonnages and reasonably allocating berth resources, and the average waiting time of ships only increases by 58 minutes.

[0049] To understand in detail how the rolling scheduling algorithm optimizes the operation efficiency of container terminals, the rolling scheduling optimization results of 35 container ships arriving within 48 hours were analyzed. The relevant information of the FCFS model and the rolling scheduling model is shown in Table 1. The order of the ships is sorted according to the best scheduling results obtained by the rolling scheduling algorithm, rather than according to their arrival times. As can be seen from Table 1, the waiting time of the rolling scheduling model is less than that of the FCFS model, especially in the ship scheduling on the second day. The reason for this phenomenon may be that the FCFS model places fairness above efficiency. Although in the ship scheduling on the first day, the waiting time of the FCFS model is not much different from that of the rolling scheduling model, small container ships in the FCFS model occupy berths that do not match their tonnage, which causes the ships entering this berth subsequently to be unable to enter the port immediately. And because the ships that arrive first have a higher priority to enter the port, the congestion is spread layer by layer, and finally the ships on the second day generally encounter congestion. However, the rolling scheduling model avoids the problem of ships occupying berth resources that do not match their tonnage too much through reasonable berth allocation. For example, S6 with a shorter operation time uses Berth 6. In addition, the rolling scheduling model also improves the channel utilization rate and further reduces the ship waiting time by swapping the arrival order of ships. For example, S14 is arranged to enter the port after S15 and S16. Since the earliest available time of the berth corresponding to S14 is 1500 minutes, and there is a channel idle time of 316 minutes from the arrival time of the previous ship entering the port at this time. Arranging S15 and S16 to enter the port during this idle time not only improves the channel utilization rate but also alleviates the congestion in the anchorage.

[0050] By looking at the data in columns 3 to 7 of Table 1, it can be seen that, for example, S17, S18, and S20 satisfy the requirement that there should be at least a 30-minute safety distance between two adjacent ships entering the channel. S3 and S14, S13 and S19 all satisfy the constraints of the "berth tandem operation" mode. This measure greatly reduces the waiting time of the following ships and alleviates the port congestion. In addition, S5 needs to enter the port at high tide. The high tide periods on that day are [360, 600] and [1080, 1320], and the arrival time of S5 is 397 minutes, which satisfies the tidal constraint. In summary, it can be concluded that the results of the rolling scheduling optimization algorithm meet the safety requirements of the theoretical model.

[0051] Table 1 Rolling Scheduling Optimization Plan for Arriving Ships (Arrival Rate: 80 min / ship)

[0052]

[0053]

[0054] In this embodiment, the basic principles, main features and advantages of the present invention have been shown and described. For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic features of the present invention, the present invention can be implemented in other specific forms.

[0055] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. The rolling dispatch optimization system for container ships in the berthing hot connection mode is characterized by: The optimization system includes model construction and rolling scheduling optimization algorithm. In the rolling scheduling optimization algorithm, when the number of ships is large, a heuristic algorithm is selected for solution, and all ships are grouped according to the arrival date. All ships arriving within one day are scheduled together, and the port entry order and berth allocation of ships arriving on the second day will be affected by the scheduling results of the ships scheduled on the previous day. The following basic assumptions are made during the model construction process to simplify the construction process of the optimization model: the berth types are all discrete berths; the estimated arrival time, operation volume and ship tonnage of the container ships arriving on the same day are known in advance; the quay crane operation efficiency of each berth is fixed; and ships sailing in the same direction cannot be overtaken.

2. The container ship rolling scheduling optimization system in the berthing hot connection mode according to claim 1 is characterized in that: In the process of model construction, based on the objectives and constraints of the scheduling in the "mooring hot connection" mode, the formula for establishing the optimization model is as follows: T b -T i ≥M(1-X ib );i∈N,b∈B (6) X ib ,F ij ,D i ,E iq ∈{0,1}(13)。 3. The container ship rolling scheduling optimization system in the berthing hot connection mode according to claim 2 is characterized by: In the optimization model formula, the ship set N = {1, 2, ... n}; the berth set B = {1, 2, ... b}; T i is the cargo capacity of ship i; U is the berth operation efficiency; t safe is the safe distance of the ship (expressed in time); is the arrival time of ship i; is the time when ship i enters the port; is the time when ship i leaves the berth; t anc It is the sailing time of the ship from the anchorage to the berth; [TS,TE] represents the traffic control period, ships are prohibited from entering the port from TS, and the traffic control ends at TE.

4. The container ship rolling scheduling optimization system in the berthing hot connection mode according to claim 2 is characterized by: T b is the upper limit of the load capacity of berth b; t tidestart is the start time of the high tide period; t interval is the interval time between each climax period; t tideend is the end time of the high tide period; Q is the set of high tide periods; M is a sufficiently large positive integer; X ib When ship i goes to berth b, it takes 1, otherwise it takes 0; F ij The value is 1 when the j ship enters the port after the i ship, otherwise it is 0; D i When ship i needs to enter the port during the tide, it takes 1, otherwise it takes 0; E iq It takes 1 when the i ship enters the port during the tidal period q, and takes 0 otherwise.

5. The container ship rolling scheduling optimization system in the berthing hot connection mode according to claim 2 is characterized in that: Formula (1) is the objective function, which means that the total waiting time of all ships is minimized; Formula (2) means that the start time of entering the port should be later than the arrival time; Formula (3) means that the time when a ship leaves the berth is the sum of its start time of entering the port, navigation time in the channel and berth operation time; Formula (4) means that according to port regulations, two adjacent ships entering the channel need to maintain a safe distance; Formula (5) is the constraint on the order of ships entering the port one after another; Formula (6) means that the ship tonnage should match the berth length.

6. The container ship rolling dispatch optimization system in the berthing hot connection mode according to claim 5 is characterized in that: Formula (7) means that large tonnage ships must enter the port during high tide; Formula (8) and Formula (9) mean that tide-riding ships must follow the time constraints of high tide; Formula (10) is the "mooring hot connection" mode constraint. When the terminal dispatching command center confirms that the remaining time for the berthing ship does not exceed the time for the berthing ship to arrive at the berth from the anchorage, it can notify the following ship to enter the port; Formulas (11)-(12) are traffic control and waterway closures caused by bad weather. No ship can enter the port during the traffic control period.

7. The container ship rolling dispatch optimization system in the berthing hot connection mode according to claim 1 is characterized by: The rolling scheduling optimization algorithm process is as follows: ship data is grouped by date; global variables are initialized; ships are processed date by date; an initial population is randomly generated, where n represents the number of convergences and is initially 0; fitness values ​​are evaluated and the best individuals are recorded; and the difference between the best fitness value of the previous generation and the best fitness value is calculated to be less than a threshold.

8. The container ship rolling dispatch optimization system in the berthing hot connection mode according to claim 7 is characterized in that: If the difference with the best fitness of the previous generation is less than the threshold, the convergence is checked by n=n+1 to see if n is greater than the convergence lower limit. If n is greater than the convergence lower limit, the optimal ship scheduling result for the day is output, the feasibility is verified, the global variables are updated, the scheduling plan for the day is accumulated, and finally the optimal scheduling plan for all ships is output.

9. The container ship rolling scheduling optimization system in the berthing hot connection mode according to claim 8 is characterized in that: If n is less than the convergence lower limit, the elite selection strategy, two-point crossover strategy, and crossover mutation strategy are used in turn to generate a new population, and the process of evaluating the fitness value and recording the optimal individual is returned.

10. The container ship rolling dispatch optimization system in the berthing hot connection mode according to claim 7, characterized in that: If the difference between the best fitness of the previous generation and that of the previous generation is greater than the threshold, n=0, and convergence is checked, and whether n is greater than the lower limit of convergence is further determined.

Images