Automatic scheduling method and device for unit

Abstract

The invention discloses an automatic scheduling method and device for a unit, and belongs to a task scheduling method.The method comprises the steps that task ring information and unit resource information are read, the task ring information comprises flight tasks and on-duty days, and the unit resource information comprises a unit member set and personnel positions corresponding to the unit member set; expanding posts for the task ring information according to personnel configuration standards in the unit resource information; the restriction rule and the constraint condition are converted into corresponding data structures, and a set coverage model is established. Manual operation is replaced with an automatic unit scheduling algorithm, cost loss caused by lack of manual service experience is reduced, scheduling efficiency and unit operation quality can be improved, better attendance experience is provided for crew personnel, and improvement of overall benefits of airline companies is assisted.

Description

TECHNICAL FIELD

[0001] The present application relates to a task scheduling method, and more particularly to a crew automatic scheduling method and device. BACKGROUND

[0002] At present, most airlines adopt manual scheduling method, that is, flights are divided according to aircraft type, base and post, and different roles of planners are responsible for the assignment of pilots. In the assignment process, the planner reasonably arranges the pilots according to the flight operation standard, combined with the qualifications, occupancy, and flying inspection requirements of the pilots, and reasonably arranges the pilots under the condition of moderately considering the balance of flight time, the balance of overnight times, and the balance of attendance days. Different flight operation standards are different, and the qualifications of pilots are also different. When manually screening, firstly, idle pilots at the base at the time or crew plans of different pilots that can be exchanged are screened out, and secondly, the qualifications and various restrictions of the pilots are checked. On this basis, the corresponding scheduling plan is manually formulated. When manually identifying, international routes, high-altitude routes, personnel flying, and overnight conditions are also considered in priority. If the personnel of the base are insufficient, the pilots of other bases may also need to be borrowed. The manual scheduling method highly depends on the personal experience of the planner, and the adjustment results and effects also vary from person to person. Especially when the number of flights increases in the peak travel season, the number of pilots is insufficient, and the routes are complex, the artificial consideration method has certain limitations and it is difficult to cover all the constraint conditions, so the scheduling quality is greatly reduced. With the rapid growth of the fleet size and the number of flights of the airline, the scarcity of crew resources is increasingly prominent, and in addition to the need for flight safety and reasonable use of crew resources, it is necessary to study and improve the crew automatic scheduling system. SUMMARY

[0003] One of the purposes of the present application is to provide a crew automatic scheduling method and device to solve the technical problems in the prior art that the manual scheduling method highly depends on the personal experience of the planner, the consistency of the adjustment results is poor, and the artificial consideration method has limitations and is difficult to cover all the constraint conditions, so the scheduling quality is poor.

[0004] To solve the above technical problems, the present application adopts the following technical solutions:

[0005] The present application provides a crew automatic scheduling method, which comprises the following steps:

[0006] Step A, reading task cycle information and crew resource information, wherein the task cycle information comprises flight tasks and duty days, and the crew resource information comprises a crew member set and corresponding personnel positions.

[0007] Step B, expanding the post according to the personnel configuration standard in the crew resource information.

[0008] Step C, converting the constraint rules and constraint conditions into corresponding data structures to establish a set cover model.

[0009]

[0010] In the above formula, M is a set of crew members to be assigned; M * is the number of personnel in the crew member set; R m is a set of all legal personal shift plans that crew member m can perform; P is a set of all task rings in the task ring information; C p is the cost of task ring p; C rm is the total cost of crew member m performing personal shift plan r; a rp is a parameter of 0-1, indicating whether task ring p is in personal shift plan r; X rm is a variable of 0-1, indicating whether crew member m performs personal shift plan r.

[0011] Step D, inputting the set cover model into a solver to solve and output an automatic scheduling result, which is each crew member in the crew resource information and a corresponding series of flight tasks.

[0012] As a preferred, a further technical solution is that the automatic scheduling result in step D is the optimal solution obtained by the solver solving the set cover model multiple times.

[0013] A further technical solution is that the solver in step D first establishes an optimization objective of the optimal solution before solving the set cover model, and the optimization objective includes any one or more of international route balance, high-altitude route balance, flight time balance, overnight times balance, and crew member attendance day balance.

[0014] A further technical solution is that the solver in step D is OR-Tools.

[0015] Another aspect of the present application provides a crew automatic scheduling device, which can execute a computer program for executing the above method.

[0016] Another aspect of the present application provides a computer readable medium, which stores instructions in the computer readable storage medium, and when the computer executes the instructions, the computer executes the above method.

[0017] Compared with the prior art, one of the beneficial effects of the present application is that a complex large-scale problem is decomposed into multiple small-scale sub-problems by an algorithm, and through parallel processing, the calculation efficiency can be significantly improved, time and resources can be saved, and a solution closer to the optimal solution can be found by gradually improving the current solution. The professional level requirement of the scheduling planner is weakened, and the relevant personnel can quickly get started. At the same time, the automatic crew scheduling algorithm replaces manual operation, reduces the cost loss caused by lack of manual business experience, improves the scheduling efficiency and crew operation quality, thereby providing better attendance experience for crew members and helping to improve the overall benefits of the airline. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A method flowchart for illustrating an embodiment of the present application.

[0019] Figure 2 An interface diagram of task ring information and crew resource information in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The principle realized by the present application is to input a task ring set, a crew member set meeting the qualification condition range, construct a feasible roster set, select the best roster (set coverage model), and crew members bid. Through a set division model, each pilot and his corresponding series of flight tasks are calculated, ensuring that each post can only be assigned to one pilot, while making the overall roster result optimal (such as: maximum task coverage rate, relatively balanced tasks for personnel of the same level, etc.). Through depth + breadth search, a feasible task combination (person + task ring / post) of all pilots within the roster period that meets the rule requirements and connection requirements is constructed, and the feasible roster combination ranked in the front of each pilot is selected through bidding. In the process of matching the crew to the task ring, the crew member qualification requirements need to be matched according to the flight operation standards, while considering the optimization goals of international routes, high-altitude routes, personnel flying, overnight conditions, etc. Based on this algorithm modeling, the optimal solution is obtained using OR-Tools, and the flight crew dispatching plan is output, realizing the overall coverage of the flight crew, while ensuring the safety of flight operation and the balance of pilots.

[0021] Based on the above-described implementation mode, the present application will be further described below in conjunction with the accompanying drawings.

[0022] One embodiment of the present application is a crew automatic scheduling method, which is executed by a computer, as shown in Figure 1 , and mainly executed according to the following steps:

[0023] Step 1, read the task ring information and crew resource information, the task ring information includes flight tasks and on-duty days, and the crew resource information includes a crew member set and corresponding personnel positions.

[0024] Step 2, expand the post according to the personnel configuration standard in the crew resource information. The form and structure of the foregoing task ring are as shown in Figure 2 It can be seen that the task contains task ring and crew configuration information, and the scheduling plan of the personnel station position includes task ring and post information. Figure 2

[0025] Step 3, convert the restriction rules and constraint conditions into corresponding data structures to establish a set covering model:

[0026]

[0027]

[0028] In the formula, M is a crew member set to be assigned; M * is the number of personnel in the crew member set; R m is a set of all legal personal scheduling plans that can be executed by crew member m; P is a set of all task rings in the task ring information; C p is the cost of task ring p; C rm is the total cost of the personal scheduling plan r executed by the crew member m; a rp The parameter is 0-1, which indicates whether the task ring p is in the personal scheduling plan r; X rm The variable is 0-1, which indicates whether the crew member m executes the personal scheduling plan r.

[0029] Step 4, input the set covering model into a solver to solve and output an automatic scheduling result, which is a series of flight tasks corresponding to each crew member in the crew resource information. In this embodiment, the foregoing automatic scheduling result is the optimal solution obtained by the solver by solving the set covering model multiple times.

[0030] In this step, OR-Tools can be used as a solver. OR-Tools is a powerful and efficient operations research optimization toolkit developed and open sourced by Google, which includes an open source integer programming solver. The present application does not include the solver related content, and is aimed at building an operations research model and then using the solver to solve, and more importantly, based on the above calculation model, the solver can use other open source or commercial solvers to achieve the same or similar purpose as the foregoing OR-Tools.

[0031] ​Based on the above method, another form of the embodiment of the application provides a computer-based device that can be used to perform the method according to the above steps. Accordingly, the method can also be stored as instructions in a computer-readable medium, so that when the computer executes the instructions, the computer executes the method of the above embodiment.

[0032] Based on the above embodiment, more specifically, in step 4, the solver first establishes an optimization objective of the optimal solution before solving the set covering model, and the optimization objective includes any one or more of international route balance, plateau route balance, flight time balance, overnight times balance, and crew member attendance days balance. Specifically, a quantitative statistical indicator (such as the number of times, hours, and days) is defined for each balance dimension (international, plateau, flight time, overnight, and attendance days). A mathematical expression for measuring the degree of imbalance is defined for these indicators, and common ones include: the variance (Variance) of the indicators between members, the difference between the maximum value and the minimum value (Range), the sum of absolute deviations (Sum of Absolute Deviations), and the like. In the optimization model, the minimization of these imbalance degree expressions is set as the optimization objective (or part of the objective).

[0033] The above balance requirements are not hard input constraints of the set covering model, but are integrated into the model as soft optimization objectives.

[0034] The integration method is to multiply the quantitative value of the “imbalance degree” corresponding to each balance objective by a weight coefficient reflecting its importance, and then add it to the main cost objective function to form a total objective function of a multi-objective weighted sum. The objective of the model is to minimize the value of the total objective function.

[0035] In the algorithm provided by the application: first, the initialization data includes: original flights, base crew resources, limit rules (121 operation limits, etc.), constraint conditions (crew strength matching, etc.). Second, based on the set division principle, personnel are assigned to task ring posts, constraints are added, a target function is constructed, a mathematical model is established, and then the OR-Tools tool is used for fast solving. In the process of task assignment, the qualification requirements of flight operation are fully considered, the balance of each task is ensured on the basis of ensuring the safe operation of flights, the fatigue degree of pilots is reduced, better attendance experience is provided for pilots, and the cost of flight operation is reduced.

[0036] The above balance target mechanism solver considers the cost and various balances when searching for a feasible solution that meets the flight coverage and regulatory hard constraints. The weight determines how to balance the cost and different balances. A high weight of an imbalance item will force the solver to find a more balanced solution, even if the cost increases slightly. The final selected 'optimal solution' is a solution that achieves the best balance between cost and various balances under the premise of meeting all hard requirements.

[0037] Compared with the manual crew dispatching mode, the crew automatic scheduling algorithm has great advantages. Especially in the peak season of flight operation, international and airport complex situations, compared with manual identification, the crew automatic scheduling algorithm can dynamically, efficiently and accurately sort out the dispatching scheme, avoid the loss caused by human factors, meet the actual operation demand and greatly improve the operation efficiency.

[0038] The flight crew scheduling problem solved by the application needs to meet complex constraints (such as crew qualification, flight time limit, rest time, flight connection, etc.), and belongs to a mixed problem of mixed integer programming (MIP) and constraint programming (CP). Through the automatic scheduling algorithm, the crew scheduling is converted into a mixed problem of mixed integer programming (MIP) and constraint programming (CP) for solving, a model is constructed, and the optimal solution is obtained by using OR-Tools.

[0039] In addition to the above, it should be noted that the "one embodiment", "another embodiment", "embodiment" and the like mentioned in the specification refer to the specific features, structures or characteristics described in conjunction with the embodiment, which are included in at least one embodiment described in the general description of the application. The same expression appears in several places in the specification does not necessarily refer to the same embodiment. Further, when a specific feature, structure or characteristic is described in conjunction with any embodiment, it is claimed that the implementation of such feature, structure or characteristic in conjunction with other embodiments also falls within the scope of the application.

[0040] Although the application is described herein with reference to a number of illustrative embodiments of the application, it should be understood that various other modifications and implementations could be devised by those skilled in the art without departing from the principles of the present disclosure. More specifically, various modifications and improvements could be made to the components of the subject combination layout and / or the layout itself without departing from the scope or spirit of the disclosure. Other uses for the components of the subject combination layout and / or the layout itself will also become apparent from the description and drawings.

Claims

1. A method of automatic generation of crew schedules, characterized by The method comprises the following steps: reading task cycle information and crew resource information, the task cycle information comprising flight tasks and on-duty days, and the crew resource information comprising a crew member set and corresponding personnel positions; expanding positions according to personnel configuration standards in the crew resource information based on the task cycle information; converting restriction rules and constraint conditions into corresponding data structures to establish a set covering model; where M is the set of crew members to be assigned; M * is the number of people in the set of crew members; R m is the set of all legal personal schedules executable by crew member m; P is the set of all tour rings in the tour ring information; C p is the cost of tour ring p; C rm is the total cost of crew member m executing personal schedule r; a rp is a parameter that is 0-1, whether tour ring p is in personal schedule r; X rm is a variable that is 0-1, whether crew member m executes personal schedule r; inputting the set covering model into a solver to solve and output an automatic scheduling result, the automatic scheduling result being a series of flight tasks corresponding to each crew member in the crew resource information.

2. The method of claim 1, wherein: The automatic scheduling result is an optimal solution obtained by the solver after multiple times of solving the set covering model.

3. The method of claim 1 or 2, wherein: Before solving the set covering model, the solver first establishes an optimization target of the optimal solution, the optimization target comprising any one or more of international route balance, plateau route balance, flight time balance, overnight times balance, and crew member on-duty day balance.

4. The method of claim 1, wherein: The solver is OR-Tools.

5. An automatic unit scheduling device, characterized by: The device can execute a computer program for executing the method in claims 1-4.

6. A computer readable medium characterized by: The computer readable storage medium stores instructions, and when the computer executes the instructions, the computer executes the method in claims 1-4.

Images