CN117391371A - Method and device for disassembling and distributing work order tasks and related medium thereof - Google Patents
Method and device for disassembling and distributing work order tasks and related medium thereof Download PDFInfo
- Publication number
- CN117391371A CN117391371A CN202311399816.1A CN202311399816A CN117391371A CN 117391371 A CN117391371 A CN 117391371A CN 202311399816 A CN202311399816 A CN 202311399816A CN 117391371 A CN117391371 A CN 117391371A
- Authority
- CN
- China
- Prior art keywords
- work order
- model
- disassembly
- langchain
- task
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000012549 training Methods 0.000 claims abstract description 121
- 238000009826 distribution Methods 0.000 claims abstract description 44
- 239000000126 substance Substances 0.000 claims abstract description 30
- 238000003889 chemical engineering Methods 0.000 claims abstract description 26
- 238000013135 deep learning Methods 0.000 claims abstract description 16
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 45
- 230000006870 function Effects 0.000 claims description 34
- 238000012423 maintenance Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 238000005457 optimization Methods 0.000 claims description 17
- 239000013598 vector Substances 0.000 claims description 17
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 10
- 230000015654 memory Effects 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000007726 management method Methods 0.000 description 22
- 230000008569 process Effects 0.000 description 11
- 230000010354 integration Effects 0.000 description 8
- 230000002787 reinforcement Effects 0.000 description 7
- 230000006872 improvement Effects 0.000 description 6
- 238000002372 labelling Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000003058 natural language processing Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000008485 antagonism Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013506 data mapping Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0633—Workflow analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/16—Real estate
- G06Q50/163—Property management
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for disassembling and distributing work order tasks and a related medium thereof, wherein the method comprises the following steps: acquiring work order training data, and preprocessing the work order training data to obtain structural chemical order text data; training a Langchain model based on a deep learning frame by combining structural chemical engineering bill text data and a preset work bill disassembly rule, and constructing to obtain a work bill disassembly model; and carrying out work order task disassembly and distribution on the work order to be disassembled by using the work order disassembly model to obtain a work order task disassembly and distribution result. According to the embodiment of the invention, the work order training data is preprocessed to obtain the structural chemical engineering order text data, then the work order training data is combined with the preset work order dismantling rule to train the Langchain model, the work order dismantling model is constructed to obtain the work order dismantling model, and finally the work order dismantling model is utilized to carry out work order task dismantling and allocation, so that the dismantling and allocation efficiency is improved, the dismantling and allocation result with higher accuracy is obtained, and the resources and cost required by the work order task dismantling and allocation are saved.
Description
Technical Field
The invention relates to the technical field of property management, in particular to a method and a device for disassembling and distributing work order tasks and a related medium thereof.
Background
The management and service work of the property industry covers a plurality of aspects such as facility management, environmental cleaning, security assurance, owner communication and the like, wherein the processing of the property work order is the key content of the management and service work of the property industry. Because the specific content in the property bill may involve multiple tasks and responsibility departments, the processing of the property bill generally generates multiple complex job bill tasks, such as facility maintenance, safety inspection, outdoor cleaning and other job bill tasks, and each job bill task may also include multiple sub-job bill tasks with different processing conditions, such as multiple sub-job bill tasks with different working positions, different working times or different responsibility persons. If the work order task is not disassembled and the processing personnel are not allocated, the problems of increased processing complexity, difficult resource allocation, difficult progress control, difficult quality control, insufficient risk management and the like of the property management and service work can be caused. The disassembly and distribution of the work order task at present usually requires staff to rely on professional knowledge and experience to disassemble and distribute manually, so that a great deal of time and labor operation and maintenance cost can be consumed, and the disassembly and distribution result is unreasonable due to negligence of the staff, thereby causing the phenomenon that resources are wasted or the work order task cannot be completed in time. Therefore, how to improve the efficiency and accuracy of work order task disassembly and distribution is a problem to be solved by those skilled in the art.
Disclosure of Invention
The embodiment of the invention provides a method, a device, computer equipment and a storage medium for disassembling and distributing work order tasks, aiming at improving the efficiency and the accuracy of work order task disassembling and distributing work.
In a first aspect, an embodiment of the present invention provides a method for disassembling and distributing work order tasks, including:
acquiring work order training data, and preprocessing the work order training data to obtain structural chemical engineering order text data;
training a Langchain model based on a deep learning frame by combining the structural chemical engineering bill text data with a preset work bill disassembly rule, and constructing to obtain a work bill disassembly model;
and carrying out work order task disassembly and distribution on the work order to be disassembled by using the work order disassembly model to obtain a work order task disassembly and distribution result.
In a second aspect, an embodiment of the present invention provides a device for disassembling and distributing work order tasks, including:
the data processing unit is used for acquiring the work order training data and preprocessing the work order training data to obtain structural chemical order text data;
the model training unit is used for training the Langchain model based on the deep learning frame by combining the structural chemical engineering bill text data with a preset work bill disassembly rule, and constructing to obtain a work bill disassembly model;
And the disassembly and distribution unit is used for disassembling and distributing the work order tasks of the work order to be disassembled by using the work order disassembly model to obtain work order task disassembly and distribution results.
In a third aspect, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the method for disassembling and allocating work order tasks according to the first aspect when executing the computer program.
In a fourth aspect, an embodiment of the present invention provides a computer readable storage medium, where a computer program is stored, where the computer program is executed by a processor to implement the method for disassembling and distributing work order tasks according to the first aspect.
The embodiment of the invention discloses a method and a device for disassembling and distributing work order tasks and a related medium thereof, wherein the method comprises the following steps: acquiring work order training data, and preprocessing the work order training data to obtain structural chemical engineering order text data; training a Langchain model based on a deep learning frame by combining the structural chemical engineering bill text data with a preset work bill disassembly rule, and constructing to obtain a work bill disassembly model; and carrying out work order task disassembly and distribution on the work order to be disassembled by using the work order disassembly model to obtain a work order task disassembly and distribution result. According to the embodiment of the invention, firstly, the work order training data is preprocessed to obtain the structural chemical order text data, then the structural chemical order text data is combined with the preset work order disassembly rules, the Langchain model based on the deep learning frame is trained, the work order disassembly model is constructed, and finally, the work order disassembly and distribution of the work order to be disassembled are carried out by utilizing the work order disassembly model, so that the work order task disassembly and distribution result with higher accuracy is obtained while the disassembly and distribution efficiency is improved, and the resources and cost required by the work order task disassembly and distribution are saved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for disassembling and distributing work order tasks according to an embodiment of the present invention;
FIG. 2 is a schematic sub-flowchart of a method for disassembling and distributing work order tasks according to an embodiment of the present invention;
FIG. 3 is a schematic block diagram of a device for disassembling and distributing work order tasks according to an embodiment of the present invention;
fig. 4 is a sub-schematic block diagram of a device for disassembling and distributing work order tasks according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Referring to fig. 1, fig. 1 is a flow chart of a method for disassembling and distributing work order tasks according to an embodiment of the present invention, which specifically includes: steps S101 to S103.
S101, acquiring work order training data, and preprocessing the work order training data to obtain structural chemical engineering order text data;
s102, training a Langchain model based on a deep learning frame by combining the structural chemical engineering bill text data with a preset work bill disassembly rule, and constructing to obtain a work bill disassembly model;
s103, the work order disassembly model is utilized to disassemble and distribute work order tasks of the work order to be disassembled, and a work order task disassembly and distribution result is obtained.
In this embodiment, first, the work order training data is acquired and preprocessed, so that the structural chemical order text data obtained by preprocessing is suitable for training, then the structural chemical order text data is combined with a preset work order disassembly rule, a Langchain model based on a deep learning frame is trained, the complexity of work order disassembly and allocation can be understood and processed after the Langchain model is trained, a work order disassembly model is constructed, finally, the work order disassembly model is utilized to disassemble and allocate work orders to be disassembled, and accordingly the work order task disassembly and allocation result with higher accuracy is obtained while the disassembly and allocation efficiency is improved, and resources and cost required by the work order task disassembly and allocation are saved.
The Langchain model in this embodiment is a deep learning model constructed based on a deep learning frame (such as a TensorFlow open source frame, a pyrerch learning library, etc.) and a large language model (such as a GPT-4 language model, a BERT language model, etc.) by a natural language processing (NLP, natural Language Processing) technology, and is used for analyzing work order text and basic disassembly of tasks in a specific application scenario. The deep learning framework provides a computing infrastructure and an algorithm library required by the Langchain model, and the large language model is a foundation of the Langchain model and provides an infrastructure and pre-trained model parameters for the Langchain model.
It should be noted that the work order disassembly model in this embodiment may be further extended and applied in the management and service work of the property industry. For example, a worksheet management system (e.g., CRM system, etc.) is built on the basis of the worksheet disassembly model, through which property management works such as worksheet analysis, fault repair processing, complaint processing, etc. can be performed.
The fault repairing process comprises the following steps: firstly, receiving property fault information input by a user, secondly, extracting fault key information such as fault types, positions and the like from the property fault information, utilizing a work order disassembly model to disassemble the fault key information into fault tasks, then distributing the fault tasks to corresponding maintenance personnel or teams through an integrated resource scheduling function, finally, acquiring maintenance progress update in real time through integration with other management systems, and providing corresponding progress feedback for the user.
The complaint treatment comprises the following steps: firstly, receiving complaint content provided by a user, secondly, extracting complaint key information from the complaint content by using a text analysis technology, and disassembling the complaint key information into a complaint task by using a work order disassembly model, then distributing the complaint task to proper processing personnel or departments, and finally, monitoring the progress of complaint processing by using a task tracking function and providing timely processing feedback for the user.
The community notification management comprises the following steps: after the work order disassembly model is used for completing disassembly and distribution of work order tasks and processing the work order tasks, the work order disassembly model is integrated with a community management system to assist in generating and managing community notification content and rules related to work order task processing results, then the community notification is sent to a user through integration with communication channels (such as message pushing, mail and short messages), then personalized customization is carried out on the community notification according to personal preferences through user attribute analysis, and finally feedback of the user is received through integration with a feedback system, and corresponding processing and replying are carried out on the feedback. When receiving information and content provided by a user, encryption and access control methods can be used for ensuring confidentiality and integrity of data and protecting privacy security of the user.
As described in connection with fig. 2, in an embodiment, the S101 includes: steps S201 to S204.
S201, acquiring initial work order training data from a work order database and a local database, and cleaning the initial work order training data to obtain candidate work order training data; the initial work order training data comprises work order content, work order solutions and historical work order disassembly records, and the local database comprises maintenance guidelines or SLA standards;
s202, acquiring a work order condition corresponding to the candidate work order training data based on the work order content; the work order conditions comprise work order emergency degree, work order type and work order problem;
s203, performing text normalization processing on the candidate work order training data to obtain the work order training data in a unified text format;
s204, converting the work order training data in the unified text format into data format to obtain the structural chemical order text data.
In this embodiment, data cleaning is performed on initial work order training data to obtain candidate work order training data, then work order conditions corresponding to the candidate work order training data are obtained based on work order content in the candidate work order training data, and then the candidate work order training data are normalized to work order training data in a unified text format.
In a specific embodiment, the worksheet training data may be represented by the dataset d= { D 1 ,d 2 ,…,d n Each work order training data D in D is represented i And all the information comprises the relevant information of worksheets such as service types, places, time, emergency degrees and the like. The preprocessing work of the work order training data is divided into data cleaning and data formatting, specifically, operations such as removing repeated data, correcting spelling errors, standardizing text formats, classifying work order contents and the like, and invalid information, error contents, inconsistency and format problems in the original work order training data can be removed through data cleaning and data formatting, so that the training quality of a Langchain model is ensured. Specifically, the worksheet training data may be preprocessed using the Pandas library in the Python computer language.
The maintenance guide in the embodiment is used for providing information such as standard operation sequence, operation requirement, notice and the like of the work order task; the SLA (Service Level Agreement ) guidelines are a contract or agreement between a service provider and a customer specifying details of what services the service provider should offer to the customer, what quality criteria the service should be offered, and how the service should be compensated for when it is not up to standard. Of course, the way or mode of acquiring the training data of the initial work order is not limited to the work order database and the local database, but text and data related to the work order can be collected by collecting property service data stored by various property management systems, analyzing the root cause of emergency (RCA) and the like, and the text and the data can be used as the training data of the initial work order. The content of the work order condition is not limited to the work order emergency degree, the work order type and the work order problem, and can also comprise the department requirement corresponding to the work order. In addition, the acquired initial work order training data can be integrated to form a corpus, so that subsequent calling is facilitated.
In another embodiment, after preprocessing the work order training data, the work order training data may be further subjected to labeling processing. The pretreatment of the work order training data can provide high-quality input data for the labeling treatment, and then the work order training data is divided into different categories (such as maintenance requests, customer complaints, facility reservation, community activities and the like) through the labeling treatment, so that data samples with labels are provided for subsequent work order disassembly and task allocation work, and the targeted treatment and tracking of the work order tasks of different categories can be carried out.
Furthermore, the labeling process can divide the work order training data through a time dimension, a space dimension and a value dimension. The time dimension is used for processing the work order problems of long-term and short-term memory of the work order training data, such as historical maintenance requests of users, property complaint records and the like; the space dimension is used for processing work order problems related to a specific place, such as the use condition of a facility at a certain position, the safety problem of a certain area such as a southern or northern city, and the like; the value dimension is used for giving value attributes and weights to worksheets with high processing efficiency and high customer satisfaction in records of worksheet processing aiming at different worksheets. On the basis of labeling processing, the Langchain model can better disassemble work order training data and distribute the work order training data to different property service personnel, such as maintenance service personnel, customer service personnel, activity organization service personnel and the like, so that the different property service personnel can concentrate on processing work orders related to responsibilities of the different property service personnel, and the efficiency and quality of property service are improved. Of course, the dimension of the labeling process partition may be other dimensions, and is not limited to the time dimension, the space dimension, and the value dimension.
In one embodiment, the step S102 includes:
preliminary randomizing parameters of the Langchain model based on a deep learning framework;
inputting the preset work order disassembly rules and the structural chemical engineering order text data into the Langch ain model to perform work order task disassembly allocation training;
evaluating the result of training output of the Langchain model by using a loss function, and carrying out minimization optimization on the loss function by using an optimization algorithm;
iteratively training the Langchain model, and updating parameters of the Langchain model until the loss function reaches a preset threshold value;
and fine tuning the work order disassembly rule, and integrating the work order disassembly rule with the Langchain model after iterative training to obtain a work order disassembly model.
In this embodiment, the parameters of the Langchain model are first randomized, then the Langchain model is trained through the work order disassembly rule and the structural chemical engineering single text data, then the Langchain model is optimized by using the loss function, and meanwhile, the Langchain model is trained for multiple rounds in an iterative mode, so that the parameters are more accurate. And then, finely adjusting a preset work order disassembly rule according to a training result of the Langchain model, and integrating the finely adjusted work order disassembly rule and the iteratively trained Langchain model into the work order disassembly model. After the training integration step is carried out, the work order disassembly model has better analysis capability and disassembly capability, so that the effects of rapidly and accurately reading, disassembling and distributing various work orders (such as maintenance facilities, cleaning public areas, arranging safety inspection and the like) are achieved, the work order tasks can be processed with high efficiency and high accuracy, and a property management enterprise can provide higher-quality service. Furthermore, the work order disassembly model can be applied to the existing property management platform, so that property management personnel can receive and process the work order in real time through the property management platform without consuming a great deal of time and cost for manual interpretation, disassembly and distribution. In a specific application scene, methods such as supervised learning or reinforcement learning can be selected to carry out work order task disassembly and allocation training.
In a specific embodiment, the work order disassembly rule in this embodiment is a set of predefined rules or criteria, which are used to analyze content and context information in the work order and instruct how to disassemble and allocate the work order, and specifically, the work order disassembly rule may be defined and set according to the urgency of the work order task, required resources, dependency relationships, and the like. For example, a work order task of "facility maintenance" may be disassembled into sub-work order tasks such as "circuit inspection" and "lamp replacement" according to specific facility types and maintenance requirements, the disassembly process may be automatically performed by a decision tree, a rule engine or other machine learning algorithm, and when the sub-work order task is disassembled, an optimization algorithm (e.g., a dynamic planning algorithm, a genetic algorithm, etc.) is used to ensure that the sub-work order tasks are performed in an optimal sequence and time schedule, and based on the disassembly rule, the work order task is allocated to an optimal resource condition and emergency according to the content of the work order, so as to automatically optimize the task execution sequence and time schedule, thereby improving the efficiency and accuracy of the work order processing.
It should be noted that the work order disassembly rule in this embodiment has different applications and purposes in the Langchain model training stage and the fine tuning integration stage. Specifically, in the Langchain model training stage, the work order disassembly rule is mainly used as a guiding factor for training the Langchain model, namely, the work order disassembly rule can be a predefined label or other form of information, and the label or the information can be embedded into an objective function (or a loss function) for training the Langchain model, so that the learning direction of the Langchain model is guided, and the Langchain model can be disassembled and distributed correctly according to the content of the work order.
In the fine tuning integration stage, the practical application environment of the work order disassembly model is considered to be more complex than the Langchain model training environment, and the condition which cannot be observed in training data may occur. Therefore, although the Langchain model already has a certain work order disassembly rule in the training stage, in order to cope with the complexity and the variability of the practical application environment, the work order disassembly rule is further defined or fine-tuned, so that the work order disassembly rule can be more focused on adapting to practical business logic and user requirements, the flexibility of the work order disassembly model is improved, and the work order disassembly model can be better adapted to different business scenes and requirements. In addition, the work order disassembly rules in the work order disassembly model can be continuously optimized and dynamically adjusted through reinforcement learning, multitask learning and other methods.
In an embodiment, the inputting the preset work order disassembly rule and the structural chemical order text data to the Langchain model to perform work order task disassembly and allocation training includes:
setting the work order disassembly rule according to the service requirement and the workflow;
based on the work order disassembly rule, mapping the structural chemical order text data by utilizing a multi-layer neural network in the Langchain model to obtain a work order task set; wherein the work order task set comprises a plurality of sub work order tasks;
The minimum task allocation cost of the sub-work order task is calculated according to the following steps:
wherein Q represents the minimum task allocation cost, n represents the total number of sub-work order tasks, m represents the total number of task personnel, c ij Representing the cost of classifying the ith sub-work order task to the jth task personnel, x ij Representation c ij Is used for determining the distribution decision variable of the system;
and confirming task personnel optimally allocated to each sub-work order task in the work order task set according to the minimum task allocation cost.
In this embodiment, a work order disassembly rule required by a Langchain model for disassembling work order tasks is set first, then structural chemical order text data is input into the Langchain model, a work order task set corresponding to the structural chemical order text data is obtained through mapping of the work order disassembly rule and a multi-layer neural network in the Langchain model, then minimum task allocation cost of sub work order tasks in the work order task set is calculated, and task personnel optimally allocated to all the sub work order tasks are confirmed by the minimum task allocation cost.
In a specific embodiment, after the data format conversion of the worksheet training data, the structured worksheet text data x=embedding (d) i ) Wherein d is i The method comprises the steps that the work order training data are represented, the Embedding represents vector conversion, structural chemical order text data in a vector form are used as starting points of multi-layer neural network processing, a work order task set T=f (X; theta) can be obtained through interaction of the multi-layer neural network and parameters theta of a Langchain model, the task set T comprises a plurality of work order tasks T, and the work order tasks T can be further subdivided into a plurality of sub-work order tasks so as to meet finer execution requirements.
Further, the mapping process of the Langchain model to the structural chemical engineering bill text data by using the multi-layer neural network can be expressed as f (Q) =t, wherein Q represents the structural chemical engineering bill text data, f () represents a function for disassembling the work bill in the Langchain model, T represents a mapped work bill task set, and all sub work bill tasks in the work bill task set comprise contents such as a work bill executing step, an executing responsibility person, an executing time and the like.
In an embodiment, the disassembling and distributing method further includes:
acquiring a work order condition corresponding to the structural chemical engineering order text data input into the Langchain model, and judging whether the work order condition reaches a preset condition threshold;
if the work order condition is judged to reach a preset condition threshold, a condition generating model is called to disassemble and distribute the structural chemical order text data;
And if the work order condition does not reach the preset condition threshold, disassembling and distributing the structural chemical order text data by utilizing the multi-layer neural network in the Langchain model.
In this embodiment, considering that under certain specific application scenarios or conditions, for example, when data in a property bill has highly irregular or complex features, special processing needs to be performed on the property bill, when the structural chemical bill text data is input into the Langchain model, a work bill condition corresponding to the structural chemical bill text data is obtained, and then the work bill condition is compared with a preset condition threshold value, so as to confirm whether special processing needs to be performed on the structural chemical bill text data. Specifically, when the condition of the work order reaches the preset condition threshold, for example, the emergency degree in the condition of the work order is high, and exceeds the preset condition threshold-medium emergency degree, the condition that the text data of the structural chemical order reach the condition of specialized treatment is indicated, so that the condition generation model can be called for carrying out disassembly and distribution treatment; when the work order condition does not reach the preset condition threshold, for example, the emergency degree in the work order condition is a medium level, and the emergency degree does not exceed the preset condition threshold-medium level emergency degree, the condition that the structural chemical engineering single text data does not reach the special processing condition is indicated, so that a condition generation model is not required to be called, and the disassembly and distribution processing can be carried out only by utilizing the multi-layer neural network in the Langchain model.
The condition generating model in this embodiment is an optional component or an extension module of the Langchain model, and may specifically be a condition deep neural network or a Condition Generating Antagonism Network (CGAN). The condition generation model is used for processing more complex or personalized work order disassembly scenes, and can generate more fine and personalized disassembly distribution results under the condition of inputting specific conditions, so that more accurate and efficient task disassembly schemes can be generated when different types or complexity work orders are faced, complex work orders are processed more flexibly, more different personalized and regional requirements are met, and the quality of property service work and customer satisfaction are further improved. Specifically, the activation condition of the condition generation model may be set according to the emergency degree, type, position or location of the work order, and other features.
In the training stage of the condition generation model, the emergency degree, type, position and other features of a specific work order are encoded into a vector form, and multi-mode data fusion is carried out with a localized data source (work order database, maintenance guide, SLA standard and the like) and work order texts, so that more comprehensive information is provided for work order task disassembly and distribution. Wherein the localized data sources are used to provide expertise associated with a particular worksheet, and the worksheet text is used to provide specific descriptions and requirements. In addition, the specific steps of the multi-mode data fusion can be as follows: the method comprises the steps of firstly encoding a work order text through a natural language processing technology to obtain a text feature vector, then splicing the feature vector corresponding to a specific work order with the text feature vector, finally encoding a localized data source, further splicing or weighted averaging with the spliced result, obtaining a comprehensive feature vector fused with all relevant information of the specific work order, and taking the comprehensive feature vector as the input of training of a condition generation model, so that the condition generation model can be disassembled and distributed more finely and individually.
In an embodiment, the inputting the preset work order disassembly rule and the structural chemical order text data to the Langchain model to perform work order task disassembly and allocation training includes:
and according to the parameters of the Langchain model, calculating to obtain conditional probability according to the following formula:
wherein P (x) i I theta) represents that the Langchain model generates an i-th class work order text x under the condition that the parameter is theta i Is Z (x) i Class i output vector, Z (x), representing a Langchain model k K-th class output vectors of the Langchain model are represented, k represents indexes of all classes of the output vectors, and theta represents parameters of the Langchain model;
and constructing and obtaining the loss function according to the following formula according to the conditional probability:
L(θ)＝-E[logP(x i |θ)]
wherein L represents the loss function and E represents a desired operation;
and calculating the gradient of the parameters of the Langchain model according to the loss function by the following formula:
wherein W represents the gradient;
updating parameters of the Langchain model according to the gradient by the following formula:
θ＝θ-aW
where a represents a learning rate.
In this embodiment, first, a Langchain model is calculated to generate an i-th type work order text x under the condition that a parameter is θ i Conditional probability P (x) i |θ), the conditional probability P (x) i I theta) is used to quantify the performance of the Langchain model and guide the optimization process of Langchain model parameters, then at the conditional probability P (x i And (theta) calculating a loss function L of the Langchain model on the basis of the I, wherein the loss function L is used for measuring the difference between a work order text generated by the Langchain model and a true value, calculating a gradient W of parameters according to the loss function L, iterating for a plurality of times according to the gradient W, continuously updating the parameter theta of the Langchain model, and obtaining the optimal parameter theta of the Langchain model by minimizing the loss function, so that the accuracy of a result generated by the Langchain model is higher, and the optimization of the Langchain model is realized.
The conditional probability P (x i I θ) is used in the formulaIs Z (x) i The index of (2) represents the unnormalized probability of judging that the work order belongs to the ith class by using the Langchain model,/>Representing all unnormalized probabilitiesk non-normalized probabilities), and the summed value varies with the input of the Langchain model and the variation of the parameters, the function of which is to normalize the non-normalized probabilities and ensure that the sum of all the non-normalized probabilities after normalization is 1.
In one embodiment, the conditional probability P (x i I θ) is output by a softmax function in the Langchain model. In addition, the loss function of the Langchain model comprises indexes such as disassembly accuracy, manual disassembly similarity and the like.
In an embodiment, the method for disassembling and distributing the work order task further includes:
collecting feedback information about work order task disassembly and distribution results, and setting a reward signal according to the feedback information;
and based on the reward signal, performing iterative optimization on parameters of the work order disassembly model by using a Q-Learning algorithm.
In this embodiment, feedback information about the work order task disassembly and distribution result output by the work order disassembly model is collected first, the feedback information can be used to understand the requirements and challenges in the work order disassembly actual operation, then a reinforcement Learning reward signal about the work order disassembly model is set according to the specific content of the feedback information, finally, based on the reward signal, the Q-Learning algorithm (the Q-Learning algorithm adopted in this embodiment is a reinforcement Learning algorithm, Q (quality) represents a function) is used to iteratively optimize the parameters of the work order disassembly model, so as to maximize positive rewards and minimize negative rewards, so that the work order task disassembly and distribution result output by the work order disassembly model can more meet the requirements of actual application occasions.
For example, user satisfaction scores or improvement suggestions for work order disassembly assignment results may be collected and converted to positive rewards and improvement suggestions to negative rewards, i.e., the objective of iterative optimization of parameters of the work order disassembly model is to increase satisfaction scores and decrease improvement suggestions. The condition for stopping the iterative optimization may be set according to actual demands, for example, it may be set that the iterative optimization is stopped when the satisfaction score reaches a preset score threshold, and the improvement suggestion is less than the preset suggestion threshold.
In addition, the quality of the disassembly and distribution results of the work order disassembly model can be evaluated by using a cost function, or the differential rewards between the work order task disassembly and distribution results and the manual disassembly and distribution results output by the work order disassembly model can be calculated, and the work order disassembly model is evaluated and optimized through the differential rewards.
In a specific embodiment, in order to collect the work order data more quickly, improve the performance and the processing effect of the work order disassembly model, and form more efficient interaction with the actual service, so that the work order disassembly model can be more flexibly adapted to different work order types and service scenes, thereby providing more accurate and efficient work order processing service, and the work order disassembly model can be optimized by adopting an RLHF (Reinforcement Learning for Human Feedback, human feedback reinforcement learning) method. The RLHF method specifically includes a multitasking learning method, a strategy gradient method, a real-time interactive training method, a personality learning method, and the like.
The multi-task learning method combines the work order task with other related tasks (such as resource allocation optimization tasks) and trains the work order disassembly model, so that the disassembly performance of the work order disassembly model in the face of different tasks is improved. In modern enterprise operation and maintenance, the work order disassembly model is required to handle various types of work order tasks, such as equipment maintenance requests, system fault reports, customer complaints, and the like. The work order disassembly rules can be adjusted according to the type and the context of the work order by the work order disassembly model after the multi-task learning reinforcement, so that the flexibility and the generalization capability of the work order disassembly model are improved. Specifically, the Multi-arm slot machine (Multi-arm bandwidth) algorithm can be used for Multi-task learning, and the basic idea of the Multi-arm slot machine (Multi-arm bandwidth) algorithm is to find a balance between exploration and utilization of a work order disassembly model.
The policy gradient method (such as Monte Carlo policy gradient algorithm) can be used for optimizing the work order disassembly rules of the work order disassembly model, so that the work order disassembly model can generate corresponding disassembly tasks according to the characteristics of the work order. In addition to the policy gradient method, a Deep Q-Network (DQN) or a depth deterministic policy gradient (Deep Deterministic Policy Gradient, DDPG) may be used to optimize the work order disassembly rules of the work order disassembly model.
The real-time interactive training method collects training data through real-time interaction with worksheet disassembly processing personnel. For example, the work order disassembly model may present problems when handling work orders, guide operators to provide feedback, and perform online training of the work order disassembly model according to the feedback, so that the model can be quickly adapted to new work order types and scenes.
The personalized learning method is used for understanding and adapting to personalized requirements of each property management enterprise or department by training a work order disassembly model, so that respective specific work order processing flows and preferences of different property management enterprises or departments are met. Specifically, the personality learning method includes:
priority identification of work order type: the work order disassembly model can optimize the work order disassembly rules through learning the priorities according to the priority setting of the property management enterprises or departments which pay more attention to equipment maintenance requests or pay more attention to system fault reports.
And (3) work order processing style adaptation: the worksheet processing style is set according to the manner in which property management enterprises or departments prefer detailed and complete worksheet disassembly or more pleasing, quick and concise processing. The work order disassembly model may learn and adapt to work order processing styles to provide more personalized work order disassembly services.
Worksheet language style matching: the work order language style is set according to the formal work order description or the expression which prefers the spoken language by the property management enterprise or department, and the work order disassembly rule is adjusted by the work order disassembly model according to different work order language styles so as to improve the accuracy and satisfaction degree of work order processing.
Fig. 3 is a schematic block diagram of an apparatus 300 for disassembling and distributing work order tasks according to the present embodiment, where the apparatus 300 includes:
the data processing unit 301 is configured to obtain work order training data, and pre-process the work order training data to obtain structural chemical order text data;
the model training unit 302 is configured to combine the structural chemical engineering bill text data with a preset work bill disassembly rule to train a Langchain model based on a deep learning frame, and construct a work bill disassembly model;
and the disassembly and distribution unit 303 is configured to disassemble and distribute the work order task of the work order to be disassembled by using the work order disassembly model, so as to obtain a work order task disassembly and distribution result.
As shown in connection with fig. 4, in an embodiment, the data processing unit 301 includes:
the data cleaning unit 401 is configured to obtain initial work order training data from a work order database and a local database, and perform data cleaning on the initial work order training data to obtain candidate work order training data; the initial work order training data comprises work order content, work order solutions and historical work order disassembly records, and the local database comprises maintenance guidelines or SLA standards;
A work order condition acquiring unit 402, configured to acquire a work order condition corresponding to the candidate work order training data based on the work order content; the work order conditions comprise work order emergency degree, work order type and work order problem;
a text normalization processing unit 403, configured to perform text normalization processing on the candidate work order training data, so as to obtain the work order training data in a unified text format;
and the data format conversion unit 404 is configured to perform data format conversion on the worksheet training data in a unified text format, so as to obtain structural chemical worksheet text data.
In one embodiment, the model training unit 302 includes:
the preliminary randomizing unit is used for carrying out preliminary randomization on parameters of the Langchain model based on the deep learning framework;
the disassembly and distribution training unit is used for inputting the preset work order disassembly rules and the structural chemical engineering order text data into the Langchain model to perform work order task disassembly and distribution training;
the minimized optimization unit is used for evaluating the result of training output of the Langchain model by using a loss function and minimizing and optimizing the loss function by using an optimization algorithm;
The iterative training unit is used for iteratively training the Langchain model and updating parameters of the Langchain model until the loss function reaches a preset threshold value;
and the fine adjustment integration unit is used for fine adjustment of the work order disassembly rule and integration with the Langchain model after iterative training to obtain a work order disassembly model.
In an embodiment, the disassembled allocation training unit comprises:
the disassembly rule setting unit is used for setting the work order disassembly rule according to the service requirement and the workflow;
the data mapping unit is used for mapping the structural chemical engineering bill text data by utilizing the multi-layer neural network in the Langchain model based on the work bill disassembly rule to obtain a work bill task set; wherein the work order task set comprises a plurality of sub work order tasks;
the allocation cost calculation unit is used for calculating the minimum task allocation cost of the sub-work order task according to the following formula:
wherein Q represents the minimum task allocation cost, n represents the total number of sub-work order tasks, m represents the total number of task personnel, c ij Representing the cost of classifying the ith sub-work order task to the jth task personnel, x ij Representation c ij Is used for determining the distribution decision variable of the system;
and the personnel allocation unit is used for confirming the task personnel optimally allocated to each sub-work order task in the work order task set according to the minimum task allocation cost.
In one embodiment, the apparatus 300 for disassembling and distributing work order tasks further includes:
the condition judging unit is used for acquiring a work order condition corresponding to the structural chemical engineering order text data input into the Langchain model and judging whether the work order condition reaches a preset condition threshold;
the condition generation model calling unit is used for calling a condition generation model to disassemble and distribute the structural chemical engineering single text data if the condition of the work sheet reaches a preset condition threshold;
and the multi-layer neural network calling unit is used for carrying out disassembly and distribution processing on the structural chemical engineering bill text data by utilizing the multi-layer neural network in the Langchain model if the work order condition is judged not to reach the preset condition threshold.
In an embodiment, the disassembled allocation training unit comprises:
the conditional probability calculation unit is used for calculating the conditional probability according to the following formula according to the parameters of the Langchain model:
wherein P (x) i I theta) represents that the Langchain model generates an i-th class work order text x under the condition that the parameter is theta i Is Z (x) i Class i output vector, Z (x), representing a Langchain model k K-th class output vectors of the Langchain model are represented, k represents indexes of all classes of the output vectors, and theta represents parameters of the Langchain model;
the loss function calculation unit is used for constructing and obtaining the loss function according to the conditional probability and the following formula:
L(θ)＝-E[logP(x i |θ)]
wherein L represents the loss function and E represents a desired operation;
the gradient calculation unit is used for calculating the gradient of the parameters of the Langchain model according to the loss function and the following formula:
wherein W represents the gradient;
the parameter updating unit is used for updating the parameters of the Langchain model according to the gradient and the following formula:
θ＝θ-aW
where a represents a learning rate.
In one embodiment, the apparatus 300 for disassembling and distributing work order tasks further includes:
the rewarding signal setting unit is used for collecting feedback information about the work order task dismantling and distributing results and setting rewarding signals according to the feedback information;
and the optimization updating unit is used for carrying out iterative optimization on the parameters of the work order disassembly model by utilizing a Q-Learning algorithm based on the reward signal.
Since the embodiments of the apparatus portion and the embodiments of the method portion correspond to each other, the embodiments of the apparatus portion are referred to the description of the embodiments of the method portion, and are not repeated herein.
The embodiment of the present invention also provides a computer readable storage medium having a computer program stored thereon, which when executed can implement the steps provided in the above embodiment. The storage medium may include: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The embodiment of the invention also provides a computer device, which can comprise a memory and a processor, wherein the memory stores a computer program, and the processor can realize the steps provided by the embodiment when calling the computer program in the memory. Of course, the computer device may also include various network interfaces, power supplies, and the like.
In the description, each embodiment is described in a progressive manner, and each embodiment is mainly described by the differences from other embodiments, so that the same similar parts among the embodiments are mutually referred. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section. It should be noted that it would be obvious to those skilled in the art that various improvements and modifications can be made to the present application without departing from the principles of the present application, and such improvements and modifications fall within the scope of the claims of the present application.
It should also be noted that in this specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. The method for disassembling and distributing the work order tasks is characterized by comprising the following steps:
acquiring work order training data, and preprocessing the work order training data to obtain structural chemical engineering order text data;
training a Langchain model based on a deep learning frame by combining the structural chemical engineering bill text data with a preset work bill disassembly rule, and constructing to obtain a work bill disassembly model;
And carrying out work order task disassembly and distribution on the work order to be disassembled by using the work order disassembly model to obtain a work order task disassembly and distribution result.
2. The method for disassembling and distributing work order tasks according to claim 1, wherein the steps of obtaining work order training data and preprocessing the work order training data to obtain structural chemical order text data comprise:
acquiring initial work order training data from a work order database and a local database, and cleaning the initial work order training data to obtain candidate work order training data; the initial work order training data comprises work order content, work order solutions and historical work order disassembly records, and the local database comprises maintenance guidelines or SLA standards;
acquiring work order conditions corresponding to the candidate work order training data based on the work order content; the work order conditions comprise work order emergency degree, work order type and work order problem;
performing text normalization processing on the candidate work order training data to obtain the work order training data in a unified text format;
and converting the work order training data in the unified text format into data format to obtain the structural chemical order text data.
3. The method for disassembling and distributing the work order task according to claim 2, wherein the training the La ngchain model based on the deep learning frame by combining the structural chemical order text data and the preset work order disassembly rule, and constructing the work order disassembly model comprises the following steps:
preliminary randomizing parameters of the Langchain model based on a deep learning framework;
inputting the preset work order disassembly rules and the structural chemical engineering order text data into the Langch ain model to perform work order task disassembly allocation training;
evaluating the result of training output of the Langchain model by using a loss function, and carrying out minimization optimization on the loss function by using an optimization algorithm;
iteratively training the Langchain model, and updating parameters of the Langchain model until the loss function reaches a preset threshold value;
and fine tuning the work order disassembly rule, and integrating the work order disassembly rule with the Langchain model after iterative training to obtain a work order disassembly model.
4. The method for disassembling and distributing work order tasks according to claim 3, wherein the step of inputting the preset work order disassembly rule and the structural chemical order text data into the Langchain model for work order task disassembly and distribution training comprises the steps of:
Setting the work order disassembly rule according to the service requirement and the workflow;
based on the work order disassembly rule, mapping the structural chemical order text data by utilizing a multi-layer neural network in the Langchain model to obtain a work order task set; wherein the work order task set comprises a plurality of sub work order tasks;
the minimum task allocation cost of the sub-work order task is calculated according to the following steps:
wherein Q represents the minimum task allocation cost, n represents the total number of sub-work order tasks, m represents the total number of task personnel, c ij Representing the cost of classifying the ith sub-work order task to the jth task personnel, x ij Representation c ij Is used for determining the distribution decision variable of the system;
and confirming task personnel optimally allocated to each sub-work order task in the work order task set according to the minimum task allocation cost.
5. The method for disassembly and distribution of work order tasks according to claim 3, further comprising:
acquiring a work order condition corresponding to the structural chemical engineering order text data input into the Langchain model, and judging whether the work order condition reaches a preset condition threshold;
if the work order condition is judged to reach a preset condition threshold, a condition generating model is called to disassemble and distribute the structural chemical order text data;
And if the work order condition does not reach the preset condition threshold, disassembling and distributing the structural chemical order text data by utilizing the multi-layer neural network in the Langchain model.
6. The method for disassembling and distributing work order tasks according to claim 3, wherein the step of inputting the preset work order disassembly rule and the structural chemical order text data into the Langchain model for work order task disassembly and distribution training comprises the steps of:
and according to the parameters of the Langchain model, calculating to obtain conditional probability according to the following formula:
wherein P (x) i I theta) represents that the Langchain model generates an i-th class work order text x under the condition that the parameter is theta i Is Z (x) i Class i output vector, Z (x), representing a Langchain model k K-th class output vectors of the Langchain model are represented, k represents indexes of all classes of the output vectors, and theta represents parameters of the Langchain model;
and constructing and obtaining the loss function according to the following formula according to the conditional probability:
L(θ)＝-E[logP(x i |θ)]
wherein L represents the loss function and E represents a desired operation;
and calculating the gradient of the parameters of the Langchain model according to the loss function by the following formula:
Wherein W represents the gradient;
updating parameters of the Langchain model according to the gradient by the following formula:
θ＝θ-aW
where a represents a learning rate.
7. The method for disassembling and assigning work order tasks according to claim 1, further comprising:
collecting feedback information about work order task disassembly and distribution results, and setting a reward signal according to the feedback information;
and based on the reward signal, performing iterative optimization on parameters of the work order disassembly model by using a Q-Learning algorithm.
8. A worksheet task disassembling and distributing device, comprising:
the data processing unit is used for acquiring the work order training data and preprocessing the work order training data to obtain structural chemical order text data;
the model training unit is used for training the Langchain model based on the deep learning frame by combining the structural chemical engineering bill text data with a preset work bill disassembly rule, and constructing to obtain a work bill disassembly model;
and the disassembly and distribution unit is used for disassembling and distributing the work order tasks of the work order to be disassembled by using the work order disassembly model to obtain work order task disassembly and distribution results.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of disassembly and allocation of work order tasks according to any of claims 1 to 7 when the computer program is executed by the processor.
10. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the computer program implements the method for disassembling and distributing work order tasks according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311399816.1A CN117391371A (en) | 2023-10-26 | 2023-10-26 | Method and device for disassembling and distributing work order tasks and related medium thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311399816.1A CN117391371A (en) | 2023-10-26 | 2023-10-26 | Method and device for disassembling and distributing work order tasks and related medium thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117391371A true CN117391371A (en) | 2024-01-12 |
Family
ID=89468143
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311399816.1A Pending CN117391371A (en) | 2023-10-26 | 2023-10-26 | Method and device for disassembling and distributing work order tasks and related medium thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117391371A (en) |
-
2023
- 2023-10-26 CN CN202311399816.1A patent/CN117391371A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Higgins et al. | Manufacturing planning and control: Beyond MRP II | |
| Lemańska-Majdzik et al. | Identification of business processes in an enterprise management | |
| US20200327467A1 (en) | Method and system for automated project management workflow and monitoring | |
| US10740679B1 (en) | Analytics scripting systems and methods | |
| US20170255999A1 (en) | Processing system to predict performance value based on assigned resource allocation | |
| Simonsson et al. | IT governance decision support using the IT organization modeling and assesment tool | |
| Elzamly et al. | Modelling and evaluating software project risks with quantitative analysis techniques in planning software development | |
| Shniekat et al. | Efficiency and effect on the competitive advantage of management information systems (MIS) in classified hotels in the city of petra; type of management as moderator | |
| Królikowski et al. | A concept of a training project IT management system | |
| Elzamly et al. | Managing software project risks (Implementation Phase) with proposed stepwise regression analysis techniques | |
| Arsanjani et al. | Avenues to improving the effectiveness of project portfolio management in the construction industry | |
| US20140297334A1 (en) | System and method for macro level strategic planning | |
| Valanarasu et al. | Risk assessment and management in enterprise resource planning by advanced system engineering theory | |
| CN117391371A (en) | Method and device for disassembling and distributing work order tasks and related medium thereof | |
| Danylevych et al. | Digitalization of Personnel Management Processes: Reserves for Using Chatbots. | |
| US20210272216A1 (en) | Severance event modeling and management system | |
| US20210334718A1 (en) | System for managing enterprise dataflows | |
| Ziegler et al. | TRUMPF uses a mixed integer model as decision support for strategic production network design | |
| US20100223101A1 (en) | Sustainability capital planning tool | |
| Aljuhani et al. | Robotic process automation and reengineering using Bizagi and UiPath: case study on mortgage request process | |
| Flechsig | The impact of intelligent process automation on purchasing and supply management–Initial insights from a multiple case study | |
| Brandt et al. | Computer-supported reuse of project management experience with an experience base | |
| Canitz | Machine Learning in Supply Chain Planning--When Art & Science Converge. | |
| Zhang et al. | An intelligent scheduling model of computer human resources in complex scenarios based on artificial intelligence | |
| US20210334753A1 (en) | System for managing enterprise dataflows |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |