CN105893727A - Spatial repulsion algorithm of resource scheduling problem on the basis of timing constraint - Google Patents
Spatial repulsion algorithm of resource scheduling problem on the basis of timing constraint Download PDFInfo
- Publication number
- CN105893727A CN105893727A CN201410741734.5A CN201410741734A CN105893727A CN 105893727 A CN105893727 A CN 105893727A CN 201410741734 A CN201410741734 A CN 201410741734A CN 105893727 A CN105893727 A CN 105893727A
- Authority
- CN
- China
- Prior art keywords
- resource
- time
- algorithm
- state space
- search
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to a limited resource scheduling problem operation method. The algorithm especially can greatly shorten the calculation time of a distribution problem of discrete resources on the basis of timing constraint. For a situation that the total quantity and the sub quantity of resource supply and consumption are known, a resource distribution scheme based on a time sequence is solved, and efficiency is quite high. A series of mutually-constrained data structures is constructed, resource distribution can be finished through the staggered arrangement of resources in the time sequence, and the search of a huge state space is reduced into the search of one row so as to eliminate complex calculation brought by the search and the backtracking of the whole state space and greatly improve algorithm efficiency.
Description
One, technical field
The invention belongs to limited resources scheduling problem operation method, this algorithm, particularly with the assignment problem of discreteness resource based on time-constrain, can significantly reduce the calculating time of this problem.
Two, background technology
Limited resources scheduling problem is frequently problem in actual applications, such as: the vehicle scheduling of transportation logistics system, automatic teacher's Course Arrangement, industrial equipment lease etc.；So-called resource scheduling based on time-constrain, refer in resource scheduling, resource provider and Resource consumers all have the feature of time mutual exclusion: i.e. may only be taken by a consumer in any one resource of same moment, Resource consumers may only consume a resource at any time simultaneously.
The problem that the steric exclusion algorithm of resource scheduling based on time-constrain intends to solve is: in the case of known resource supply and the total amount of consumption and component, how resource is distributed to Resource consumers？When the distribution of these resources is not limited by partial order, but when being limited by seasonal effect in time series constraint, in real work in the case of resource provisioning kind and quantity, consuming entity enormous amount, we, how by a kind of computational methods, are found a given resource for the basic allocative decision of predetermined consumption by computer software.This is a thing the most significant, because after obtaining a basic scheme, we can be optimized based on the program and rearrange, and finally gives one closest to the scheme required.
This class problem is applied and extensive, especially all the more so in computer information system.Relate to the content of the aspect such as operational research and Combinational Mathematics on this problem theory, application further relates to computer software and data structure, algorithm design, the content of the aspects such as artificial intelligence；Solve this problem the most in practice and there is certain difficulty.Common software design arts, uses operational research thought the most fully, is confined to some angle considered a problem.And it is engaged in the experts and scholars of job scheduling research, and more focus on the leading edge in professional field and difficult point, the application for some reality but has less attention, and this is perhaps the reason that this problem does not has preferable solution always.
For the resource scheduling of time-constrain, the outer most dispatching algorithms of Present Domestic are all based on the state space depth-first search scheme of band backtracking；In arrangement spatially, constantly sound out and recall, the state space of whole arrangement is carried out exhaustive；The optimization principles of its algorithm is also based on the optimization of search volume, for example with α-β pruning strategy.These algorithms, in complicated applied environment, when state space is the biggest, still have the biggest time complexity；Even if using various optimized algorithm, it is also difficult to fundamentally reduce amount of calculation, this is for only wanting to find an application solved, and the cost of calculating is the biggest.
If being divided into two steps to consider the problems referred to above, first finding a basic satisfactory allocative decision, being then optimized in this scheme, then can dissolve complicated and time-consuming search work.
The present invention can be used to realize the first step, and use additive method to carry out arrangement optimization.
The present invention does not include how the method for Optimized Operation, and these methods can be by some classical ways of artificial intelligence field, such as: genetic algorithm etc..
Three, summary of the invention
In order to reduce a large amount of calculating that the huge state space search of resource scheduling based on time-constrain brings, this algorithm is by constructing the MRP space of a mutual exclusion, avoid the exhaustive search to state space, scheduling process is made without recall on a large scale, thus improves computational efficiency significantly.
So-called steric exclusion algorithm is through the time series to resource provisioning person and quantity delivered, construct a big two-dimensional state space, by the resource to resources consumption entity Heterogeneous Permutation sequentially in time, thus it is finally completed the distribution to consuming entity of the whole resource.The present invention is to deep search process to single resource of the search procedure boil down to of whole state space, and not backtracking, thus increases substantially the speed of distribution.
Whole computational methods are divided into two steps by the present invention, the following is several agreements of statement algorithm
{ }: the entity of expression definition, the essential feature of content representation entity, can be according to actual application extension substance feature
[]: presentation-entity set
→: represent the dependence between element
Concrete numerical procedure is as follows:
First, definition need to participate in the various data structures of scheduling
1, the data structure of definition expression basic resource unit:
Resitem{resID, resTotal}；①
Resitem: Resource Unit resID: resource id resTotal: the quantity that this resource can provide
Set up and represent resource collection: resitem [] resitems；
2, the data structure of resource consumption entity is defined: ex_resitem{ex_resID} is 2.
Ex_resitem: resource consumption entity ex_resID: resource consumption entity ID
Set up resource consumption entity sets: ex_resitem [] ex_resitems；
3, the resource consumption entity data structure to the demand relation of various resources is set up:
Rtoer:{ex_resID → [resID, resNumber] }；③
ResID: with 1. middle explanation
ResNumber: represent resource requirement quantity
[resID, resNumber]: represent the set of resID, resNumber
Each consumption entity of rtoer relational representation is for the demand of each resource
4, based on time series division stock number two-dimensional state space: resMap_t:{resitem, t1-n}；④
ResMap_t needs according to time series layout, all of resource is become bivariate table, and the behavior resID of this table is classified as the time quantum sequence of minimum essential requirement, and cross point is the stock number provided
t1-n: time quantum sequence
The most commonly known condition, by the most 3., adds each resource distribution situation in time series, it is possible to obtain resMap_t data structure.
5, the definition resources consumption entity based on the time series division data structure to the relation of resource requirement:
Ex_resMap_t:{ex_resitem, t1-n}；⑤
Consuming entity is carried out record in the resources consumption amount of each time quantum by ex_resMap_t.This relation deposits the result that the present invention asks for.Therefore, during initialization, this relation is empty.
The most 5. the time series of data structure should be consistent, thus provides safeguard for the following algorithm of enforcement.
Secondly, computational methods are realized according to below step:
1, resitem, resitems, ex_resitem, ex_resitems, rtoer, resMap_t, ex_resMap_t are initialized.
2, taking out an entity from ex_resitems, obtain required resID and resNumber according to this entity id from rtoer, if ex_resitems takes, prompting is assigned information, turns 6, otherwise continues.
3, check whether the corresponding resID of resitems has allowable resource, if there is, resMap_t is turned to find resource id to be distributed, principle (the time-constrain then only taken by a consumer at same time quantum according to a resID, steric exclusion method), the time series of this resource scans for distribution, completes the resource distribution of this consuming entity；Labelling (representing that this resource of this time quantum is unavailable) has been distributed on making on the corresponding time point of resMap_t relation after distribution；Reduce the total resources of this resID simultaneously；Continue next step.If resMap_t No Assets on corresponding component can divide, prompting distribution failure information, turn 6.
4, record and export this consumption physical resource assignment data
5, step 2 is returned to
6, terminate
By above-mentioned algorithmic procedure it can be seen that this computational methods are to complete an allocative decision according to time series, the program possesses the basic feature of various allocative decision.For a scheduling of resource project, it is understood that there may be multiple allocative decisions；Some of which scheme may more tally with the actual situation, and other may not be well adapted for working condition.But any scheme all possesses identical feature, therefore finding optimum allocation can carry out scheme optimization on the basis of the present invention, and this is not belonging to scope of the invention.
Four, accompanying drawing explanation
Fig. 1 is the building-block of logic of whole algorithm, and it has equivalent effect with above-described algorithmic procedure, will easily facilitate this algorithm of explanation with reference to reading mutually.
Fig. 2 is that in case study on implementation, part searches for a class course solves tree.
Five, detailed description of the invention
In order to make the technical problem to be solved, technical scheme and beneficial effect clearer, below in conjunction with accompanying drawing and the specific embodiment of a course arrangement, the present invention will be described in detail.It should be noted that, specific embodiment described herein is only in order to explain the present invention, it is not intended to limit the present invention.
This example uses B to represent that class, C represent that course, P represent teacher
Assuming that every class can arrange the curriculum weekly is 10 times, every teacher's maximum row's class amount weekly is 8 times, P (Ci) represents teacher's number (every teacher's correspondence a branch of instruction in school) of wantonly i-th subject, the quantity of teachers of every a branch of instruction in school is (namely course amount): P (C1)=2 P (C2)=2 P (C3)=2 P (C4)=1, thus obtain the teacher numbering relevant to course: P11, P12, P21, P22, P31, P32, P41.Wherein P11, P12 represent the sequence number of the teacher of C1 course of teaching, and other labels are analogized；Therefore course with the corresponding relation of teacher is:
C1 → (P11, P12) C2 → (P21, P22) C3 → (P31, P32) C4 → (P41)
Assuming without under particular case, the course that each teacher serves as, all by approximation mean allocation (if being not desired to mean allocation, can adjust), thus can be calculated the teacher number of 7 teacher's entities and each entity and quantity of teaching:
(P11,8) (P12,7) (P21,8) (P22,8) (P31,7) (P32,5) (P41,8)
Each class is shown in Table 1 for the demand schedule of course:
Table 1 class is to each course demand table
Setting up teacher and arrange record, be shown in Table 2, T1-T10 represents 10 courses of every class, and this table is initialized as sky.
Start algorithm: with class's number circulation, according to required course to teachers structure takes teacher (because teacher and course have corresponding relation), when the course of some teacher is 0, then take the next bit teacher of same course of teaching；Owing to teacher same on same string can not occur twice, therefore the algorithm (as a example by B2) in one's own profession course arrangement can be expressed as: needed for B2 course as C2 (2), C3 (4), C4 (2), P21 (2), P31 (4), P41 (2) is can be taken off by course teacher's mapping table and teachers data structure, find tri-row of P21, P31, P41 the most in table 2, thus this three row is set up the search tree of such as Fig. 2 (seeing accompanying drawing).Then this three row can be carried out depth-first search by programming time, the course arrangement of this class can be found out；After so completing, amendment table 2 and teachers data structure again.Mode like this, until the course arrangement of all classes completes.
Table 2 teacher's arrangement of time record
| T1 | T2 | T3 | T4 | T5 | T6 | T7 | T8 | T9 | T10 | |
| P11 | 1 | 1 | 1 | 1 | ||||||
| P12 | ||||||||||
| P21 | ||||||||||
| P22 | ||||||||||
| P31 | 1 | 1 | 1 | |||||||
| P32 | ||||||||||
| P41 | 1 | 1 |
By above-mentioned algorithm, available construction of curriculum table 3:
Table 3 construction of curriculum table
The foregoing is only a case study on implementation of the present invention, be not limited to the present invention, any amendment, equivalent and the improvement etc. made within all principles in the present invention and spirit, within being all included in protection scope of the present invention.
Claims (2)
1. the steric exclusion algorithm of resource scheduling based on time-constrain, by basic resource unit, resource consumption
Entity, consumption entity are to the demand of various resources, based on time series division stock number two-dimensional state space
The data structure of several mutual constraints and allocation algorithm are constituted.
2. according to the steric exclusion algorithm of the resource scheduling based on time-constrain described in claims 1, its
It is characterised by constructing and divides stock number two-dimensional state space resMap_t and resource in the time based on time series
Heterogeneous Permutation method in sequence.Divide stock number two-dimensional state space resMap_t based on time series to guarantee
During finding allocative decision, search volume is compressed on a row；Resource mistake in time series
Position aligning method ensure that feasibility of imputation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410741734.5A CN105893727A (en) | 2014-12-08 | 2014-12-08 | Spatial repulsion algorithm of resource scheduling problem on the basis of timing constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410741734.5A CN105893727A (en) | 2014-12-08 | 2014-12-08 | Spatial repulsion algorithm of resource scheduling problem on the basis of timing constraint |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105893727A true CN105893727A (en) | 2016-08-24 |
Family
ID=56699576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410741734.5A Pending CN105893727A (en) | 2014-12-08 | 2014-12-08 | Spatial repulsion algorithm of resource scheduling problem on the basis of timing constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105893727A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271407A (en) * | 2008-05-13 | 2008-09-24 | 武汉理工大学 | Gridding scheduling method based on energy optimization |
| CN101777146A (en) * | 2009-12-29 | 2010-07-14 | 大唐软件技术股份有限公司 | Method and device for construction scheduling |
| CN103617472A (en) * | 2013-07-09 | 2014-03-05 | 成都希盟泰克科技发展有限公司 | Resource balancing self-adaption scheduling method of multi-project and multi-task management |
-
2014
- 2014-12-08 CN CN201410741734.5A patent/CN105893727A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271407A (en) * | 2008-05-13 | 2008-09-24 | 武汉理工大学 | Gridding scheduling method based on energy optimization |
| CN101777146A (en) * | 2009-12-29 | 2010-07-14 | 大唐软件技术股份有限公司 | Method and device for construction scheduling |
| CN103617472A (en) * | 2013-07-09 | 2014-03-05 | 成都希盟泰克科技发展有限公司 | Resource balancing self-adaption scheduling method of multi-project and multi-task management |
Non-Patent Citations (5)
| Title |
|---|
| 官春平: "多资源约束作业车间调度问题的模型研究", 《广东轻工职业技术学院学报》 * |
| 崔琳等: "高校排课系统中的模拟手工排课算法分析", 《安徽电子信息职业技术学院学报》 * |
| 施步青等: "基于时间约束的Ad Hoc网格能量优化算法", 《武汉理工大学学报（交通科学与工程版）》 * |
| 李坤等: "多集装箱堆场空间分配与车辆调度集成问题的建模与优化", 《系统工程理论与实践》 * |
| 高明芳等: "高校自动排课系统排课算法的研究与设计", 《电脑知识与技术》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN207895435U (en) | Neural computing module | |
| Li et al. | An effective shuffled frog-leaping algorithm for multi-objective flexible job shop scheduling problems | |
| Hamzadayi et al. | A genetic algorithm based approach for simultaneously balancing and sequencing of mixed-model U-lines with parallel workstations and zoning constraints | |
| Bagheri et al. | An artificial immune algorithm for the flexible job-shop scheduling problem | |
| Tuncel et al. | Two-sided assembly line balancing using teaching–learning based optimization algorithm | |
| Mohammadi et al. | Development, application, and comparison of hybrid meta-heuristics for urban land-use allocation optimization: Tabu search, genetic, GRASP, and simulated annealing algorithms | |
| CN114327844A (en) | Memory allocation method, related device and computer readable storage medium | |
| CN104166903B (en) | Mission planning method and system based on process division | |
| Upasani et al. | Automated generation of dimensioned rectangular floorplans | |
| RU2012127578A (en) | ANALYZER OF APPLICABILITY OF THE SOFTWARE MODULE FOR THE DEVELOPMENT AND TESTING OF THE SOFTWARE FOR MULTIPROCESSOR MEDIA | |
| CN103488537A (en) | Method and device for executing data ETL (Extraction, Transformation and Loading) | |
| Moussa et al. | Master assembly network for alternative assembly sequences | |
| CN103235743A (en) | Method for scheduling multi-target testing task based on decomposition and optimal solution following strategies | |
| CN111915011B (en) | Single-amplitude quantum computing simulation method | |
| Dou et al. | A new hybrid algorithm for cold chain logistics distribution center location problem | |
| Sitek et al. | Hybrid solution framework for supply chain problems | |
| CN103729694A (en) | Method for solving flexible job-shop scheduling problem with improved GA based on polychromatic set hierarchical structure | |
| CN102831102A (en) | Method and system for carrying out matrix product operation on computer cluster | |
| CN100531070C (en) | Network resource scheduling simulation system | |
| CN106228235A (en) | A kind of land utilization space Optimal Configuration Method taking general layout and PROCESS COUPLING effect into account | |
| CN105404611A (en) | Matrix model based multi-calculation-engine automatic selection method | |
| CN102855278B (en) | A kind of emulation mode and system | |
| Hu et al. | A greedy-based crow search algorithm for semiconductor final testing scheduling problem | |
| CN105893727A (en) | Spatial repulsion algorithm of resource scheduling problem on the basis of timing constraint | |
| Ball et al. | Material compatibility constraints for make-to-order production planning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Wu Xujun Inventor after: Mou Chunxiao Inventor after: Zhang Zhaohui Inventor before: Wu Xujun Inventor before: Xu Baohua Inventor before: Zhang Fuyuan |
|
| CB03 | Change of inventor or designer information | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170627 Address after: 264003 Yantai, Laishan Economic Development Zone, West Road, Mingda, No. 11, No. Applicant after: Shandong Ocean Technology Co., Ltd. Address before: 264003 Yantai, Laishan Economic Development Zone, West Road, Mingda, No. 11, No. Applicant before: YANTAI HUITONG NETWORK TECHNOLOGY CO., LTD. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160824 |
|
| RJ01 | Rejection of invention patent application after publication |