CN116362599B

CN116362599B - Quality data acquisition method and device based on MES system

Info

Publication number: CN116362599B
Application number: CN202310266967.3A
Authority: CN
Inventors: 张锐; 钱孝峰; 张瑜; 刘赠
Original assignee: Wuhan Tongjie Information Technology Co ltd
Current assignee: Wuhan Tongjie Information Technology Co ltd
Priority date: 2022-12-12
Filing date: 2023-03-20
Publication date: 2023-11-10
Anticipated expiration: 2043-03-20
Also published as: CN116362599A

Abstract

The invention provides a quality data acquisition method and device based on an MES system, wherein the method comprises the following steps: generating a first odor data set and a second odor data set; spraying gas to staff in a preparation area; gas data acquisition is carried out on the whole staff to obtain acquisition data; inputting the first smell data set, the second smell data set and the acquired data into a pre-trained operation quality prediction model; removing the first smell and the second smell on the staff; and taking all staff operation quality conclusions as quality data. The invention has the beneficial effects that: on the premise that cameras are not allowed to be arranged, different gases are sprayed at different positions according to different staff, then the gas concentration is detected, and finally, a working quality prediction model based on a neural network model is used for prediction, so that the acquisition of quality data is assisted.

Description

Quality data acquisition method and device based on MES system

Technical Field

The invention relates to the field of data acquisition, in particular to a quality data acquisition method and device based on an MES system.

Background

The MES system refers to a manufacturing execution system (manufacturing execution system, called MES for short) and aims to help enterprises to realize production plan management, production process control, product quality management, workshop inventory management, project bulletin board management and the like, so that the manufacturing execution capacity of the enterprises is improved. Generally, MES systems include inspection modules for inspecting the product being produced, which inspection generally only detects the quality of the product, but lacks detection of the quality of the job by staff, so that the result of the inspection is not comprehensive, and it is difficult to quickly trace the cause of quality problems, especially in the case of large defective products.

To solve this problem, cameras may be arranged in the work area to detect the quality of work on the part of staff. However, due to various aspects such as privacy, more and more work areas are not allowed to be provided with cameras, and therefore, how to use the MES system for comprehensive quality data collection is still a problem.

Disclosure of Invention

The invention mainly aims to provide a quality data acquisition method and device based on an MES system, and aims to solve the problem that comprehensive quality data acquisition is difficult to carry out when the MES system is not allowed to arrange cameras.

The invention provides a quality data acquisition method based on an MES system, wherein the MES system comprises a planning module, a dispatching module, a production execution module and a checking module, the method is applied to the checking module, and the method comprises the following steps:

s1, generating a first smell data set and a second smell data set according to a preset smell data generation method; the first smell data set comprises a plurality of first positions, a first smell category and concentrations respectively corresponding to the first positions, and the second smell data set comprises a plurality of second positions, a second smell category and concentrations respectively corresponding to the second positions; the first position is located on the body of the natural person, and the second position is located on the hand or leg of the natural person;

s2, using a preset gas injector, and performing gas injection on staff in a preparation area according to the first smell data set and the second smell data set, so that a first position and a second position on the staff are respectively injected with a first smell and a second smell with corresponding concentrations, and then allowing the staff to enter a working area; wherein the preparation area is adjacent to the work area;

s3, when a working period is finished, gas data acquisition is carried out on the whole staff by using an odor collector arranged in a cleaning area so as to obtain acquisition data; wherein the collected data comprises at least the current first odor concentration of all first locations and the current second odor concentration of all second locations on the employee; the cleaning area is adjacent to the working area;

S4, inputting the first smell data set, the second smell data set and the acquired data into a pre-trained operation quality prediction model to obtain an operation prediction result generated by the operation quality prediction model; the operation quality prediction model is obtained by training a preset deep neural network model by adopting sample data acquired in advance; the operation prediction result comprises normal operation quality or abnormal operation quality;

s5, removing first smell and second smell on the staff by adopting preset smell removing equipment, and allowing the staff to leave the working area;

s6, sequentially executing the steps S1-S5 on other staff to obtain a plurality of operation prediction results, and taking all staff operation quality conclusions as quality data, thereby completing a quality data acquisition flow.

Further, the planning module is used for indicating to generate a production plan according to the pre-input basic data;

the dispatching module is used for indicating to dispatch the production task to the corresponding operation area according to the production plan;

the production execution module is used for instructing staff to execute production tasks so as to generate products;

the checking module is used for indicating to check the working quality and/or the product quality of staff;

The camera is not present in the working area.

Further, after step S6 of completing the quality data collection flow, the steps S1-S5 are sequentially executed on other staff members to obtain a plurality of job prediction results, and all the staff member job quality conclusions are used as quality data, including:

s61, detecting the produced product by adopting a preset product quality detection method to obtain a product quality detection result;

s62, mapping the product quality detection result into a product score according to a preset score mapping method, and mapping the quality data into a job score;

s63, calling preset weight parameters, and carrying out weight summation on the product scores and the job scores according to the weight parameters to obtain a quality total score;

s64, judging whether the total quality score is larger than a preset score threshold value or not;

s65, if the total quality score is larger than a preset score threshold, judging that the quality of the current working period is qualified;

and S66, if the total quality score is not greater than a preset score threshold, judging that the quality of the current working period is unqualified.

Further, when the operation quality prediction model processes the training process, the operation area is provided with a camera, after the operation quality prediction model is trained, the operation area does not have a camera, and before the step S4 of inputting the first smell data set, the second smell data set and the collected data into the operation quality prediction model obtained by pre-training to obtain the operation prediction result generated by the operation quality prediction model, the operation quality prediction model comprises:

S41, generating a third odor data set and a fourth odor data set according to a preset odor data generation method; the third odor data set comprises a plurality of third positions, third odor categories and concentrations respectively corresponding to the third positions, and the fourth odor data set comprises a plurality of fourth positions, fourth odor categories and concentrations respectively corresponding to the fourth positions; the third position is the same as the first position, and the fourth position is the same as the second position;

s42, using a preset gas injector, performing gas injection on the sample staff in the preparation area according to the third odor data set and the fourth odor data set, so that a third position and a fourth position on the sample staff are respectively injected with a third odor and a fourth odor with corresponding concentrations, and allowing the sample staff to enter the operation area;

s43, when a working period is finished, gas data acquisition is carried out on the whole body of a sample employee by using an odor collector arranged in a cleaning area so as to obtain sample odor data; in the same time period, adopting a preset camera to respectively perform continuous image acquisition processing on sample staff to obtain an image data sequence;

S44, combining the third odor data set, the fourth odor data set and the sample odor data into data to be marked, and marking the data to be marked according to the image data sequence to generate sample data; the labeling processing refers to labeling a label with normal operation quality or abnormal operation quality;

s45, sequentially executing steps S41-S44 on other sample staff to obtain a plurality of sample data, and dividing the plurality of sample data into a plurality of training data and a plurality of verification data according to a preset proportion;

s46, training the deep neural network model by adopting a gradient descent method according to a plurality of training data to obtain a working quality prediction model; the method comprises the steps of updating parameters of each layer of neural network in a deep neural network model by adopting a back propagation algorithm during model training;

and S47, verifying the temporary operation quality prediction model according to a plurality of verification data, and generating an operation prediction instruction on the premise that the verification result is qualified, so as to instruct the first smell data set, the second smell data set and the acquisition data to be input into the operation quality prediction model obtained by training in advance, so as to obtain an operation prediction result generated by the operation quality prediction model.

Further, the staff wears a plurality of textiles on the body, the textiles are respectively positioned on the hands or the legs, and the textiles are not contacted with each other; the first smell and the second smell are the smell which cannot be distinguished by natural people; the first smell data set and the second smell data set corresponding to different staff are different; the step S2 of using a preset gas injector to perform gas injection on the staff in the preparation area based on the first smell data set and the second smell data set, so that the first position and the second position on the staff are respectively injected with the first smell and the second smell with corresponding concentrations, includes:

s201, spraying the first textile fabric by using a preset gas sprayer and taking air as a gas source;

s202, judging whether the position of the first textile is a first position or not;

s203, if the position of the first textile is the first position, switching the air source to air with first smell, and enabling the first textile to be sprayed with the first smell with corresponding concentration according to the first smell data set;

s204, if the position of the first textile is not the first position, continuing to spray air;

S205, executing steps S201-S204 on other textiles until the last textile is sprayed by using a preset gas sprayer.

The invention also provides a quality data acquisition device based on the MES system, wherein the MES system comprises a planning module, a dispatching module, a production execution module and a checking module, the device is applied to the checking module, and the device comprises:

the odor data generation unit is used for instructing to execute the step S1 and generating a first odor data set and a second odor data set according to a preset odor data generation method; the first smell data set comprises a plurality of first positions, a first smell category and concentrations respectively corresponding to the first positions, and the second smell data set comprises a plurality of second positions, a second smell category and concentrations respectively corresponding to the second positions; the first position is located on the body of the natural person, and the second position is located on the hand or leg of the natural person;

the gas spraying unit is used for indicating to execute the step S2 and using a preset gas sprayer to spray gas to staff in a preparation area according to the first smell data set and the second smell data set, so that a first position and a second position on the staff are respectively sprayed with a first smell and a second smell with corresponding concentrations, and the staff is allowed to enter a working area; wherein the preparation area is adjacent to the work area;

The odor acquisition unit is used for indicating to execute the step S3, and acquiring gas data of the whole body of staff by using an odor collector arranged in the cleaning area when one working period is finished so as to obtain acquisition data; wherein the collected data comprises at least the current first odor concentration of all first locations and the current second odor concentration of all second locations on the employee; the cleaning area is adjacent to the working area;

the prediction result generation unit is used for indicating and executing the step S4, inputting the first smell data set, the second smell data set and the acquired data into a pre-trained work quality prediction model so as to obtain a work prediction result generated by the work quality prediction model; the operation quality prediction model is obtained by training a preset deep neural network model by adopting sample data acquired in advance; the operation prediction result comprises normal operation quality or abnormal operation quality;

an odor removing unit for instructing to execute step S5, removing the first odor and the second odor on the staff member by using a preset odor removing apparatus, and allowing the staff member to leave the work area;

the quality data acquisition unit is used for indicating to execute the step S6, sequentially executing the steps S1-S5 on other staff to obtain a plurality of operation prediction results, and taking all staff operation quality conclusions as quality data, thereby completing a quality data acquisition flow.

the camera is not present in the working area.

The invention has the beneficial effects that: on the premise that cameras are not allowed to be arranged, different gases are sprayed at different positions according to different staff, then the gas concentration is detected, finally, a working quality prediction model based on a neural network model is used for prediction, and prediction results of all the staff are integrated, so that quality data acquisition can be assisted.

Drawings

FIG. 1 is a schematic diagram of an MES system according to one embodiment of the present invention;

FIG. 2 is a flow chart of a MES-based quality data collection method according to an embodiment of the present invention;

FIG. 3 is a schematic block diagram of a quality data acquisition device based on an MES system according to an embodiment of the present invention.

The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

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 only 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 noted that, in the embodiments of the present invention, all directional indicators (such as up, down, left, right, front, and back) are merely used to explain the relative positional relationship, movement conditions, and the like between the components in a specific posture (as shown in the drawings), if the specific posture is changed, the directional indicators correspondingly change, and the connection may be a direct connection or an indirect connection.

The term "and/or" is herein merely an association relation describing an associated object, meaning that there may be three relations, e.g., a and B, may represent: a exists alone, A and B exist together, and B exists alone.

Furthermore, descriptions such as those referred to as "first," "second," and the like, are provided for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implying an order of magnitude of the indicated technical features in the present disclosure. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.

Referring to FIG. 1, the MES system of the present invention is shown as comprising a planning module, a dispatch module, a production execution module and a verification module, wherein the planning module is used for indicating to generate a production plan according to pre-input basic data; the dispatching module is used for indicating to dispatch the production task to the corresponding operation area according to the production plan; the production execution module is used for instructing staff to execute production tasks so as to generate products; the checking module is used for indicating to check the working quality and/or the product quality of staff; most importantly, the working area of the invention is free of cameras. In the figure, three dispatching points, three working areas and three inspection points are taken as examples, and the functions of each module are described, but in practical application, the number of dispatching points, the number of working areas and the number of inspection points are not limited.

Referring to fig. 2, the invention provides a quality data acquisition method based on an MES system, wherein the MES system comprises a planning module, a dispatching module, a production execution module and a checking module, the method is applied to the checking module, and the method comprises the following steps:

The invention can collect and predict the working quality data of staff without allowing cameras to be arranged, and the reasons are that the staff have different activities and different moving modes according to different work types, wherein the moving modes refer to that part of work types need to use upper arms more, other work types need to run more, therefore, legs need to be used more, and the trunk is relatively static. The difference of the activity modes can lead to the difference of the diffusion rates of the odors on the surface of the human body, so that the odor concentration change of different positions can be detected to be used as an analysis basis of the working quality of the staff, namely, the odor concentration of the staff has a clear corresponding relation with the working quality, and the deep neural network model is exactly the model for analyzing and predicting the clear corresponding relation, so that the working quality of the staff can be predicted by using the working quality prediction model.

As described in the above steps S1-S2, according to a preset smell data generating method, a first smell data set and a second smell data set are generated; the first smell data set comprises a plurality of first positions, a first smell category and concentrations respectively corresponding to the first positions, and the second smell data set comprises a plurality of second positions, a second smell category and concentrations respectively corresponding to the second positions; the first position is located on the body of the natural person, and the second position is located on the hand or leg of the natural person; using a preset gas injector, and carrying out gas injection on staff in a preparation area according to the first smell data set and the second smell data set, so that a first position and a second position on the staff are respectively injected with a first smell and a second smell with corresponding concentrations, and then allowing the staff to enter a working area; wherein the preparation area is adjacent to the work area.

The employee of the present invention may be a natural person of any identity, for example, a formal employee, or a temporary visitor. The preset odor data generation method may be any feasible method, for example, may be a random generation method or other methods, for example, may be set as follows: the staff of the same work species adopt the identical first smell data set and the second smell data set; alternatively, it may be set so that: and B, the first smell data set and the second smell data set of any staff are not identical with the first smell data set and the second smell data set of the other staff. The different settings are suitable for different scenes, the A setting is adopted, the rapid analysis with large granularity is convenient to perform, the B setting is adopted, for example, the analysis with small granularity is performed, the speed is low, but the novel effect is brought, namely, the smell data set of the staff can also be used as the mark of the staff, the specific time of entering the working area and leaving the working area of the staff can be determined, and the staff cannot imitate the smell data set only when the smell data set is kept secret, so that the data accuracy is ensured. The present invention preferably employs the B setting.

The odor injector may be connected to only two air sources or to more air sources, in the present invention, the odor injector is preferably connected to three air sources, the first air source corresponding to the first odor, the second air source corresponding to the second odor, and the third air source corresponding to air. The employee preferably wears a porous textile and, for consistency, preferably provides a uniform textile.

Further, the odor imparting of the present invention may be accomplished with any feasible gas, which may include a gas that is perceivable by a natural person or a gas that is not perceivable, preferably a gas that is not perceivable by a natural person. Wherein, although the smell is set as the gas which can not be perceived by natural people is the preferable choice of the invention, other effects can be realized if the smell is set as the smell which can be perceived by natural people, such as setting as certain fragrances, refreshing smell and the like, which is beneficial to improving the experience of staff.

The odor collector of the present invention is, for example, an electronic nose, so the category of odors can be any feasible category, but needs to be detectable by the electronic nose, and the types of gases that can be detected by the electronic nose include, but are not limited to: carbon monoxide, bromine gas, chlorine gas, unsymmetrical dimethylhydrazine, chlorine dioxide, fluorine gas, phosgene, hydrogen chloride, hydrogen fluoride, hydrogen cyanide, hydrogen sulfide, hydrogen peroxide, hydrogen, oxygen, nitrogen monoxide, iodine gas, ozone, ammonia, nitrogen dioxide, sulfur dioxide, acid gas, arsenic hydride, germane, hydrogen selenide, phosphine, silane, ethylene oxide, formaldehyde, toluene, ethanol, acetylene; alkanes, hydrocarbons, alcohols, etc., of course, in the actual selection process, selection of gases harmful to the human body should be avoided. The first odor type and the second odor type may be formed of a single gas or a mixed gas.

As described in the above step S3, at the end of one working cycle, gas data is collected on the whole body of the staff by using the odor collector disposed in the cleaning area to obtain collected data; wherein the collected data comprises at least the current first odor concentration of all first locations and the current second odor concentration of all second locations on the employee; the cleaning region is adjacent to the working region.

The cleaning zone is adjacent to the work zone and personnel leaving the work zone should pass through the cleaning zone. The odor collector is, for example, an electronic nose. The electronic nose is a kind of smell sensing device for detecting different smells, including smells that a human body can perceive and smells that cannot perceive. The scent data set is associated with the electronic nose, so the requirements of the scent data set include: the distance between the locations in the scent data set cannot be so close that the electronic nose misjudges the scent location. The collected data at least comprises the current first smell concentration of all first positions and the current second smell concentration of all second positions on the staff.

Inputting the first smell data set, the second smell data set and the collected data into a pre-trained operation quality prediction model to obtain an operation prediction result generated by the operation quality prediction model as described in the step S4; the operation quality prediction model is obtained by training a preset deep neural network model by adopting sample data acquired in advance; the job prediction result includes a job quality normal or a job quality abnormal. Because the first smell data set and the second smell data set contain position data and initial concentration data, the acquired data contain position data and final concentration data, and the division of different work species is directly related to the change value of the concentration data, the work quality of one employee is related to the total amount of activities which the work species belongs to and are supposed to execute, and the total amount of activities consists of a plurality of decomposed actions, each action can influence the gas diffusion at the corresponding position, so that the acquired data of the same employee are different in the state of high work quality and low work quality; staff of different work species do not necessarily have the same collected data under the same working quality. However, in any case, the first smell data set, the second smell data set and the collected data are directly related to the working quality of the corresponding staff, and the deep neural network model predicts the exact relation, so the invention adopts the deep neural network model to conduct working prediction. The deep neural network model adopted by the invention can be any feasible model, such as a deep convolutional neural network model, an countermeasure network model, a residual network model, a deep neural network model corrected based on a genetic algorithm, a long-term and short-term memory network model and the like, and the invention is not limited to the deep neural network model.

In addition, the invention can be realized by adopting a plurality of modes for deploying a plurality of small models, namely, different models are trained for staff with different responsibilities, so that in actual implementation, only the corresponding small model is required to be invoked for the targeted staff.

As described in the above step S5, the preset odor removing device is used to remove the first odor and the second odor on the staff member, and allow the staff member to leave the work area so as not to influence the normal activities of the staff member. The odor removing device is, for example, a high-power exhaust fan, a dust remover, etc., and is used for removing odor molecules in the previous steps.

As described in step S6 above, steps S1-S5 are sequentially performed on other employees to obtain a plurality of operation prediction results, and all employee operation quality conclusions are used as quality data, so as to complete the quality data collection flow.

In one embodiment, the steps S1-S5 are sequentially performed on other employees to obtain a plurality of job prediction results, and all job quality conclusions of the employees are taken as quality data, so that after step S6 of the quality data collection flow is completed, the method includes:

The quality data of the invention is actually auxiliary data, and because the quality data of the MES system is composed of both product quality data and employee operation quality data, the product quality detection result should be considered when generating a quality report. The product quality detection method can be any feasible method, which is different according to different product types, and a nondestructive detection method is selected. The score mapping method is to present the product quality detection result and quality data in the form of scores. And then, according to different weight parameters, carrying out weight summation on the product scores and the job scores to obtain a quality total score.

In one embodiment, when the operation quality prediction model processes the training process, the operation area is provided with a camera, after the operation quality prediction model is trained, the operation area does not have a camera, and before step S4 of inputting the first smell data set, the second smell data set and the collected data into the operation quality prediction model obtained by pre-training to obtain an operation prediction result generated by the operation quality prediction model, the operation quality prediction model comprises:

Therefore, training of the model is achieved, and a work quality prediction model is obtained, wherein collection of sample data is similar to the formal data collected in the formal implementation step. Labeling the sample data can be realized by adopting a manual labeling or automatic labeling mode, but labeling labels are all dependent on the performance of sample staff, for example, the sample staff is idle in a time period, and then an abnormal label of the activity state is given; otherwise, the active state should be given a normal label. The predetermined ratio is, for example, 0.8:0.2, 0.9:0.1, 0.95:0.05, etc. The back propagation algorithm is an algorithm based on a gradient descent method, and is suitable for training a multi-layer neural network. If the result of Korean verification using the plurality of verification data of the same source is verification pass, it indicates that the temporary activity state prediction model can be qualified as the activity state prediction task of the present invention, and thus the temporary activity state prediction model is referred to as a final activity state prediction model.

In one embodiment, the staff wears a plurality of textiles on the staff, the textiles are respectively positioned on the hands or the legs, and the textiles are not contacted with each other; the first smell and the second smell are the smell which cannot be distinguished by natural people; the first smell data set and the second smell data set corresponding to different staff are different; the step S2 of using a preset gas injector to perform gas injection on the staff in the preparation area based on the first smell data set and the second smell data set, so that the first position and the second position on the staff are respectively injected with the first smell and the second smell with corresponding concentrations, includes:

The textile is, for example, a porous fiber textile, which has a strong odor retention property and whose concentration also becomes smaller as the surface air flow rate increases. In addition, when the gas injector is used for spraying smell, the smell cannot directly go to the accurate first position for spraying, because staff can know the accurate first position by doing so. Similarly, when the odor data is collected, all textiles are comprehensively collected so as not to expose the first position.

Referring to fig. 3, the invention further provides a quality data acquisition device based on an MES system, the MES system comprises a planning module, a dispatching module, a production execution module and a checking module, the device is applied to the checking module, and the device comprises:

an odor data generating unit 10 for instructing to execute step S1, generating a first odor data set and a second odor data set according to a preset odor data generating method; the first smell data set comprises a plurality of first positions, a first smell category and concentrations respectively corresponding to the first positions, and the second smell data set comprises a plurality of second positions, a second smell category and concentrations respectively corresponding to the second positions; the first position is located on the body of the natural person, and the second position is located on the hand or leg of the natural person;

A gas spraying unit 20, configured to instruct to perform step S2, using a preset gas sprayer, and based on the first smell data set and the second smell data set, perform gas spraying on the staff in the preparation area, so that the first position and the second position on the staff are sprayed with the first smell and the second smell with corresponding concentrations, respectively, and allow the staff to enter the work area; wherein the preparation area is adjacent to the work area;

the odor collection unit 30 is configured to instruct to perform step S3, and perform gas data collection on the whole body of the employee to obtain collection data by using the odor collector disposed in the cleaning area at the end of one working cycle; wherein the collected data comprises at least the current first odor concentration of all first locations and the current second odor concentration of all second locations on the employee; the cleaning area is adjacent to the working area;

a prediction result generating unit 40, configured to instruct to perform step S4, input the first smell data set, the second smell data set, and the collected data into a task quality prediction model that is trained in advance, so as to obtain a task prediction result generated by the task quality prediction model; the operation quality prediction model is obtained by training a preset deep neural network model by adopting sample data acquired in advance; the operation prediction result comprises normal operation quality or abnormal operation quality;

An odor removing unit 50 for instructing to execute step S5, removing the first odor and the second odor on the staff member using a preset odor removing apparatus, and allowing the staff member to leave the work area;

the quality data collection unit 60 is configured to instruct to execute step S6, and sequentially execute steps S1-S5 on other staff members to obtain a plurality of operation prediction results, and take all the staff member operation quality conclusions as quality data, thereby completing a quality data collection flow.

In one embodiment, the planning module is configured to instruct generation of a production plan based on pre-entered base data;

the camera is not present in the working area.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by hardware associated with a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium provided by the present invention and used in embodiments may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method 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, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.

The embodiment of the invention can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A quality data acquisition method based on an MES system, wherein the MES system comprises a planning module, a dispatch module, a production execution module and a test module, the method is applied to the test module, and the method comprises the following steps:

s1, generating a first smell data set and a second smell data set according to a preset smell data generation method; the first smell data set comprises a plurality of first positions, a first smell category and concentrations respectively corresponding to the first positions, and the second smell data set comprises a plurality of second positions, a second smell category and concentrations respectively corresponding to the second positions; the first position is located on the body of the natural person, and the second position is located on the hand or leg of the natural person; the staff wears a plurality of textiles on the body, the textiles are respectively positioned on the hands or the legs, and the textiles are not contacted with each other;

S3, when a working period is finished, gas data acquisition is carried out on the whole staff by using an odor collector arranged in a cleaning area so as to obtain acquisition data; the collected data at least comprise the current first odor concentration of all the first positions and the current second odor concentration of all the second positions on the staff, so that the odor change amount of the first odor at the first positions before and after the end of a working period is different from the odor change amount of the second odor at the second positions before and after the end of the working period according to the principle that the different activity modes can lead to the difference of the odor diffusion rates of the surfaces of human bodies; the cleaning area is adjacent to the working area;

2. The MES-based quality data collection method according to claim 1, wherein,

the planning module is used for indicating to generate a production plan according to the basic data input in advance;

the camera is not present in the working area.

3. The MES system-based quality data collection method according to claim 2, wherein after step S6 of completing the quality data collection flow, the steps S1 to S5 are sequentially performed on other employees to obtain a plurality of operation prediction results, and all the employee operation quality conclusions are used as quality data, and the steps include:

4. The MES system-based quality data collection method according to claim 1, wherein, when the operation quality prediction model processes the training process, a camera is provided in the operation area, and after the operation quality prediction model is trained, the camera is not present in the operation area, and before step S4 of inputting the first smell data set, the second smell data set, and the collection data into the operation quality prediction model obtained by training in advance to obtain an operation prediction result generated by the operation quality prediction model, the method includes:

5. The MES system based quality data collection method of claim 1, wherein the first odor and the second odor are both odors indistinguishable by a natural person; the first smell data set and the second smell data set corresponding to different staff are different; the step S2 of using a preset gas injector to perform gas injection on the staff in the preparation area based on the first smell data set and the second smell data set, so that the first position and the second position on the staff are respectively injected with the first smell and the second smell with corresponding concentrations, includes:

6. A quality data acquisition device based on an MES system, wherein the MES system comprises a planning module, a dispatching module, a production execution module and a test module, the device is applied to the test module, and the device comprises:

the odor data generation unit is used for instructing to execute the step S1 and generating a first odor data set and a second odor data set according to a preset odor data generation method; the first smell data set comprises a plurality of first positions, a first smell category and concentrations respectively corresponding to the first positions, and the second smell data set comprises a plurality of second positions, a second smell category and concentrations respectively corresponding to the second positions; the first position is located on the body of the natural person, and the second position is located on the hand or leg of the natural person; the staff wears a plurality of textiles on the body, the textiles are respectively positioned on the hands or the legs, and the textiles are not contacted with each other;

The odor acquisition unit is used for indicating to execute the step S3, and acquiring gas data of the whole body of staff by using an odor collector arranged in the cleaning area when one working period is finished so as to obtain acquisition data; the collected data at least comprise the current first odor concentration of all the first positions and the current second odor concentration of all the second positions on the staff, so that the odor change amount of the first odor at the first positions before and after the end of a working period is different from the odor change amount of the second odor at the second positions before and after the end of the working period according to the principle that the different activity modes can lead to the difference of the odor diffusion rates of the surfaces of human bodies; the cleaning area is adjacent to the working area;

7. The MES-based quality data collection device of claim 6, wherein,

the camera is not present in the working area.

8. The MES system-based quality data collection apparatus according to claim 7, wherein after step S6 of completing the quality data collection process, the steps S1 to S5 are sequentially performed on other employees to obtain a plurality of operation prediction results, and all the employee operation quality conclusions are used as quality data, and the steps include:

9. The MES system-based quality data collection device according to claim 6, wherein, when the operation quality prediction model processes the training process, the operation area is provided with a camera, and after the operation quality prediction model is trained, the operation area does not have a camera, and before the step S4 of inputting the first smell data set, the second smell data set and the collection data into the operation quality prediction model obtained by training in advance to obtain an operation prediction result generated by the operation quality prediction model, the method comprises:

10. The MES system based quality data collection apparatus of claim 6, wherein the first odor and the second odor are both odors indistinguishable by a natural person; the first smell data set and the second smell data set corresponding to different staff are different; the step S2 of using a preset gas injector to perform gas injection on the staff in the preparation area based on the first smell data set and the second smell data set, so that the first position and the second position on the staff are respectively injected with the first smell and the second smell with corresponding concentrations, includes: