A manufacturing service combination optimization method and system based on Apriori algorithm

Abstract

The present invention discloses a manufacturing service combination optimization method and system based on the Apriori algorithm. The system comprises: a screening module for frequent manufacturing services, which first determines the minimum support and minimum confidence of the Apriori algorithm, and uses the Apriori algorithm to traverse the database to obtain a frequent item set; a manufacturing service association strength calculation module, which traverses the frequent item set and calculates the association strength of the manufacturing service; a selection module for the best combined service execution path, which generates a combined service execution path, compares the average value of the association strength of the neighboring order correlation in each combined service execution path, and obtains the best combined service execution path. The present invention screens the optimal manufacturing service combination according to the association strength between manufacturing services, and determines the best service combination execution path.

Description

Technical Field

[0001] The present invention belongs to the technical field of manufacturing service combination optimization, and specifically relates to a manufacturing service combination optimization method and system based on an Apriori algorithm. Background Art

[0002] With the rapid development of the manufacturing industry, driven by personalized and diversified market demands, manufacturing tasks have become increasingly complex. The rapid growth in the number of manufacturing services has led to an exponential growth in the number of manufacturing service combinations. Better combination of manufacturing services or selection of matching manufacturing services has become a hot issue. Constrained by problems such as insufficient production capacity, single or simple manufacturing services cannot solve the complex collaborative manufacturing tasks in intelligent manufacturing. Existing manufacturing service combination methods mainly focus on the study of adjacent-order related service combinations, without considering the global structural characteristics of multi-order composite service combinations, and cannot effectively solve the complex collaborative manufacturing tasks in manufacturing. Summary of the invention

[0003] The purpose of the present invention is to provide a manufacturing service combination optimization method and system based on the Apriori algorithm, the system includes a frequent item set generation module, a service combination association strength calculation module, and an optimal combination service execution path module, which is suitable for solving multi-order related service combination problems. The method of the present invention takes into account the global structural characteristics (connectivity, integrity, redundancy and logical correctness) of the composite service execution path of non-adjacent order correlation, analyzes the non-uniform correlation transfer from adjacent order correlation to non-adjacent order correlation, and can effectively improve the composite service combination success rate.

[0004] The technical solution of the present invention is a manufacturing service combination optimization method based on the Apriori algorithm, comprising the following steps:

[0005] Step (1) Screening of frequent manufacturing services, specifically including:

[0006] ① First, determine the minimum support minsup and minimum confidence minconf of the Apriori algorithm;

[0007] ② Use the Apriori algorithm to traverse the manufacturing service database and determine whether the support of the candidate service is greater than or equal to the minimum support;

[0008] ③ Generate a frequent manufacturing service set from frequent manufacturing services whose support is greater than or equal to the minimum support;

[0009] Step (2) Calculation of the association strength of manufacturing services, specifically including:

[0010] ① Traverse the frequent manufacturing service set and determine whether there is interdependence between the previous and next services through the confidence of the service combination;

[0011] ② Calculate the association strength of the manufacturing service combinations that meet the condition requirements, then determine the manufacturing service combination with the maximum association strength, and generate a combined service execution path for the manufacturing service combination with the maximum association strength;

[0012] Step (3) selection of the optimal manufacturing service combination execution path, specifically including:

[0013] ① Compare the average values ​​of the correlation strength of the neighboring order correlations in each composite service execution path;

[0014] ② Select the combined service with the largest average association strength to obtain the best combined service execution path.

[0015] Furthermore, the step 1 (1) specifically includes:

[0016] RS i With RS j is the manufacturing service, i, j are the serial numbers of the manufacturing services, and x is the manufacturing service RS i The number of times it appears in the manufacturing service combination execution path set, where n represents the number of all manufacturing services, is defined as the manufacturing service RS i The support of is: s(i) = x / n;

[0017] Confidence RS stands for Manufacturing Service j Manufacturing Services RS i The probability of appearing as a pre- and post-order service in the service composition execution path is expressed as follows:

[0018]

[0019] Among them, |<RS i ,RS j >| means including <RS i ,RS j >The number of service composition execution paths, |RS i |Indicates that manufacturing services are included RS i The number of service composition execution paths, |RS j |Indicates that manufacturing services are included RS j The number of execution paths of the service combination.

[0020] Furthermore, step 1 (2) specifically includes:

[0021] According to the calculation formula of manufacturing service support s(i)=x / n, the support of each manufacturing service in the manufacturing service database is calculated to determine whether its support is greater than or equal to the minimum support, that is, to determine whether s(i)≥minsup.

[0022] Further, the step 2 (2) specifically includes: according to the definition of strong correlation, medium strong correlation, medium weak correlation, if correlation; as follows:

[0023] Strong correlation:

[0024] Moderately strong correlation:

[0025] Moderately weak correlation:

[0026] Weak correlation:

[0027] Furthermore, the step 2 (2) specifically includes:

[0028] Manufacturing Services RS i and RS j The preceding correlation strength is defined as r s (RS i →RS j ) and will manufacture services RS i and RS j The post-order correlation strength is defined as r s (RS i ←RS j ); If two manufacturing services RS i and RS j If there is a strong correlation between the two manufacturing service RS i and RS j There is a moderately strong correlation between them, then Rel(i, j) = 0.8; if two manufacturing service RS i and RS j There is a medium weak correlation between them, then Rel(i, j) = 0.6; if two manufacturing service RS i and RS j There is a weak correlation between them, then Rel(i, j) = 0.2; if RS i and RS j There is no correlation between them, then Rel(i, j) = 0;

[0029] r s (RS i →RS j )=Rel(i,j)s(i,j)c(i→j) I

[0030] r s (RS i ←RS j )=Rel(i,j)s(i,j)c(i←j) II

[0031] Rel(i,j) is RS i and RS j The composable relationship between them is represented by s(i,j) to represent the manufacturing service RS i and RS j Support, Manufacturing Services RS i and RS j The correlation strength of is obtained by formula I and II, which is r s (RS i ,RS j ).

[0032]

[0033] Furthermore, the step 3 (2) specifically includes:

[0034] Calculate the average value of the correlation strength of the manufacturing service combination execution path corresponding to the four correlations.

[0035]

[0036] N represents the number of manufacturing services in the manufacturing service combination execution paths corresponding to the four types of correlations.

[0037] The present invention also proposes a manufacturing service combination optimization system based on the Apriori algorithm, comprising:

[0038] (1) A screening module for frequent manufacturing services, used to:

[0039] ① Determine the minimum support minsup and minimum confidence minconf of the Apriori algorithm;

[0040] ② Use the Apriori algorithm to traverse the manufacturing service database and determine whether the support of the candidate service is greater than or equal to the minimum support;

[0041] ③ Generate a frequent manufacturing service set from frequent manufacturing services whose support is greater than or equal to the minimum support;

[0042] (2) The module for calculating the association strength of manufacturing services is used to:

[0043] ① Traverse the frequent manufacturing service set and determine whether there is interdependence between the previous and subsequent services through the confidence of the service combination.

[0044] ② Calculate the association strength of the manufacturing service combinations that meet the condition requirements, then determine the manufacturing service combination with the maximum association strength, and generate a combined service execution path for the manufacturing service combination with the maximum association strength.

[0045] (3) A module for selecting the optimal manufacturing service combination execution path, used to:

[0046] ① Compare the average values ​​of the correlation strength of the neighboring order correlations in each composite service execution path;

[0047] ② Select the combined service with the largest average association strength to obtain the best combined service execution path.

[0048] The advantages of the present invention compared with the prior art are:

[0049] (1) Based on the neighbor-order correlation, the present invention considers the global structural characteristics of non-neighbor-order correlation in the composite service execution path, and can effectively screen out manufacturing service combinations with higher correlation strength.

[0050] (2) The present invention combines the Apriori principle in manufacturing service combination and prunes some infrequent item sets and their supersets to reduce the complexity of generating frequent item sets. As the number of candidate services in the candidate service set increases, this method has more advantages than other methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of the present invention.

[0053] According to one embodiment of the present invention, a manufacturing service combination optimization method based on the Apriori algorithm is proposed, which takes into account the global structural characteristics (connectivity, integrity, redundancy and logical correctness) of the non-neighboring order related composite service execution path. The non-uniform correlation transfer from neighboring order correlation to non-neighboring order correlation is analyzed, and based on workflow and modular design, a manufacturing service combination optimization method based on frequent item sets is proposed to perceive composable services. This method can effectively improve the success rate of multi-order related service combination matching in manufacturing service combination. Specifically, it includes the following steps:

[0054] Step (1) screening of frequent manufacturing services is specifically implemented as follows:

[0055] ①First, determine the minimum support (minsup) and minimum confidence (mincof) of the Apriori algorithm.

[0056] RS i With RS jis the manufacturing service, i, j are the serial numbers of the manufacturing services, and x is the manufacturing service RS i The number of times it appears in the manufacturing service combination execution path set, where n represents the number of all manufacturing services, is defined as the manufacturing service RS i The support degree of is: s(i) = x / n.

[0057] Confidence RS stands for Manufacturing Service j Manufacturing Services RS i The probability of appearing as a pre- and post-order service in the service composition execution path is expressed as follows:

[0058]

[0059] Among them, |<RS i ,RS j >| means including <RS i ,RS j >The number of service composition execution paths, |RS i |Indicates that manufacturing services are included RS i The number of service composition execution paths, |RS j |Indicates that manufacturing services are included RS j The number of execution paths of the service combination.

[0060] ② According to the calculation formula of manufacturing service support s(i)=x / n, calculate the support of each manufacturing service in the manufacturing service database and determine whether its support is greater than or equal to the minimum support, that is, determine whether s(i)≥min sup.

[0061] ③ Generate a frequent manufacturing service set from frequent manufacturing services that are greater than or equal to the minimum support.

[0062] (2) Calculation of the association strength of manufacturing services is specifically implemented as follows:

[0063] ① Traverse the frequent manufacturing service set and use the confidence formula (1) in the frequent manufacturing service screening module to calculate the confidence of the manufacturing services in the frequent manufacturing service set. After obtaining the confidence of the manufacturing services in the frequent manufacturing service set, use the judgment formula in Table 1 to determine whether there is a strong correlation, a medium strong correlation, a medium weak correlation, or a weak correlation.

[0064] Table 1 Combinable correlation relationships based on confidence

[0065]

[0066] ② Based on the judgment in the previous step, filter out the frequent manufacturing services that meet the four correlations, calculate the correlation strength of these frequent manufacturing services, and then determine the pre- and post-sequence manufacturing services with the maximum correlation strength.

[0067] Manufacturing Services RS i and RS j The preceding correlation strength is defined as r s (RS i →RS j ) and will manufacture services RS i and RS j The post-order correlation strength is defined as r s (RS i ←RS j ). If two manufacturing services RS i and RS j If there is a strong correlation between the two manufacturing service RS i and RS j There is a moderately strong correlation between them, then Rel(i, j) = 0.8; if two manufacturing service RS i and RS j There is a medium weak correlation between them, then Rel(i, j) = 0.6; if two manufacturing service RS i and RS j There is a weak correlation between them, then Rel(i, j) = 0.2. If RS i and RS j If there is no correlation between them, Rel(i, j)=0.

[0068] r s (RS i →RS j )=Rel(i,j)s(i,j)c(i→j) I

[0069] r s (RS i ←RS j )=Rel(i,j)s(i,j)c(i←j) II

[0070] Rel(i,j) is RS i and RS j The composable relationship between them. Use s(i,j) to represent the manufacturing service RS i and RS j support.

[0071] Manufacturing Services RS i and RS j The correlation strength can be obtained through formulas I and II, that is, r s (RSi ,RS j ).

[0072]

[0073] (3) The selection of the optimal manufacturing service combination execution path is specifically implemented as follows:

[0074] ① Combine the manufacturing services with the maximum correlation strength corresponding to the four correlations to generate the manufacturing service combination execution path. Starting from the first item of the manufacturing service, combine the first item with the service with the maximum correlation strength in the next service until the last step of the manufacturing service to generate the service execution path.

[0075] ② Calculate the average correlation strength of the manufacturing service combination execution paths corresponding to the four correlations.

[0076]

[0077] N represents the number of manufacturing services in the manufacturing service combination execution paths corresponding to the four types of correlations.

[0078] ③ Compare the average values ​​of the correlation strengths of the manufacturing service combination execution paths corresponding to the four correlations, and select the combination service with the largest average value of correlation strength as the optimal manufacturing service combination execution path.

[0079] According to another aspect of the present invention, Figure 1 As shown in the figure, a manufacturing service combination optimization system based on Apriori algorithm is also proposed, including:

[0080] (1) A screening module for frequent manufacturing services, used to:

[0081] ① Determine the minimum support minsup and minimum confidence minconf of the Apriori algorithm;

[0082] ② Use the Apriori algorithm to traverse the manufacturing service database and determine whether the support of the candidate service is greater than or equal to the minimum support;

[0083] ③ Generate a frequent manufacturing service set from frequent manufacturing services whose support is greater than or equal to the minimum support;

[0084] (2) The module for calculating the association strength of manufacturing services is used to:

[0085] ① Traverse the frequent manufacturing service set and determine whether there is interdependence between the previous and subsequent services through the confidence of the service combination.

[0086] ② Calculate the association strength of the manufacturing service combinations that meet the condition requirements, then determine the manufacturing service combination with the maximum association strength, and generate a combined service execution path for the manufacturing service combination with the maximum association strength.

[0087] (3) A module for selecting the optimal manufacturing service combination execution path, used to:

[0088] ① Compare the average values ​​of the correlation strength of the neighboring order correlations in each composite service execution path;

[0089] ② Select the combined service with the largest average association strength to obtain the best combined service execution path.

[0090] In summary, the present invention discloses a manufacturing service combination optimization method and system based on the Apriori algorithm. The system includes a candidate service combination generation module, a frequent item set generation module, a service combination association strength calculation module, and an optimal combination service execution path module. The present invention selects the optimal manufacturing service combination according to the association strength between manufacturing services and determines the optimal service combination execution path.

[0091] The contents not described in detail in the specification of the present invention belong to the prior art known to the professional and technical personnel in this field.

[0092] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A manufacturing service combination optimization method based on Apriori algorithm, characterized in that: The steps include: Step (1) Screening of frequent manufacturing services, specifically including: ① First, determine the minimum support minsup and minimum confidence minconf of the Apriori algorithm; ② Use the Apriori algorithm to traverse the manufacturing service database and determine whether the support of the candidate service is greater than or equal to the minimum support; ③ Generate a frequent manufacturing service set from frequent manufacturing services whose support is greater than or equal to the minimum support; Step (2) Calculation of the association strength of manufacturing services, specifically including: ① Traverse the frequent manufacturing service set and determine whether there is interdependence between the previous and next services through the confidence of the service combination; ② Calculate the association strength of the manufacturing service combinations that meet the condition requirements, then determine the manufacturing service combination with the maximum association strength, and generate a combined service execution path for the manufacturing service combination with the maximum association strength; Step (3) selection of the optimal manufacturing service combination execution path, specifically including: ① Compare the average values ​​of the correlation strength of the neighboring order correlations in each composite service execution path; ② Select the combined service with the largest average association strength to obtain the best combined service execution path.

2. The manufacturing service combination optimization method based on the Apriori algorithm according to claim 1 is characterized in that: The step 1① specifically includes: RS i With RS j is the manufacturing service, i, j are the serial numbers of the manufacturing services, and x is the manufacturing service RS i The number of times it appears in the manufacturing service combination execution path set, where n represents the number of all manufacturing services, is defined as the manufacturing service RS i The support of is: s(i) = x / n; Confidence RS stands for Manufacturing Service j Manufacturing Services RS i The probability of appearing as a pre- and post-order service in the service composition execution path is expressed as follows: Among them, |<RS i ,RS j >| means including <RS i ,RS j >The number of service composition execution paths, |RS i |Indicates that manufacturing services are included RS i The number of service composition execution paths, |RS j |Indicates that manufacturing services are included RS j The number of execution paths of the service combination.

3. The manufacturing service combination optimization method based on the Apriori algorithm according to claim 1 is characterized in that: The step 1 (2) specifically includes: According to the calculation formula of manufacturing service support s(i)=x / n, the support of each manufacturing service in the manufacturing service database is calculated to determine whether its support is greater than or equal to the minimum support, that is, to determine whether s(i)≥minsup.

4. The manufacturing service combination optimization method based on the Apriori algorithm according to claim 3 is characterized in that: The step 2 (2) specifically includes: strong correlation, moderately strong correlation, moderately weak correlation, and if correlation according to the definition; as follows: Strong correlation: Moderately strong correlation: Moderately weak correlation: Weak correlation:

5. The manufacturing service combination optimization method based on Apriori algorithm according to claim 1 is characterized in that: The step 2 (2) specifically includes: Manufacturing Services RS i and RS j The preceding correlation strength is defined as r s (RS i →RS j ) and will manufacture services RS i and RS j The post-order correlation strength is defined as r s (RS i ←RS j ); If two manufacturing services RS i and RS j If there is a strong correlation between the two manufacturing service RS i and RS j There is a moderately strong correlation between them, then Rel(i, j) = 0.8; if two manufacturing service RS i and RS j There is a medium weak correlation between them, then Rel(i, j) = 0.6; if two manufacturing service RS i and RS j There is a weak correlation between them, then Rel(i, j) = 0.2; if RS i and RS j There is no correlation between them, then Rel(i, j) = 0; r s (RS i →RS j )=Rel(i,j)s(i,j)c(i→j) I r s (RS i ←RS j )=Rel(i,j)s(i,j)c(i←j) II Rel(i,j) is RS i and RS j The composable relationship between them is represented by s(i,j) to represent the manufacturing service RS i and RS j Support, Manufacturing Services RS i and RS j The correlation strength of is obtained by formula I and II, which is r s (RS i ,RS j ); 6. The manufacturing service combination optimization method based on Apriori algorithm according to claim 1 is characterized in that: The step 3 (2) specifically includes: Calculate the average value of the correlation strength of the manufacturing service combination execution path corresponding to the four correlations. N represents the number of manufacturing services in the manufacturing service combination execution paths corresponding to the four types of correlations.

7. A manufacturing service combination optimization system based on Apriori algorithm, characterized in that: include: (1) A screening module for frequent manufacturing services, used to: ① Determine the minimum support minsup and minimum confidence minconf of the Apriori algorithm; ② Use the Apriori algorithm to traverse the manufacturing service database and determine whether the support of the candidate service is greater than or equal to the minimum support; ③ Generate a frequent manufacturing service set from frequent manufacturing services whose support is greater than or equal to the minimum support; (2) The module for calculating the association strength of manufacturing services is used to: ① Traverse the frequent manufacturing service set and determine whether there is interdependence between the previous and next services through the confidence of the service combination; ② Calculate the association strength of the manufacturing service combinations that meet the condition requirements, then determine the manufacturing service combination with the maximum association strength, and generate a combined service execution path for the manufacturing service combination with the maximum association strength; (3) A module for selecting the optimal manufacturing service combination execution path, used to: ① Compare the average values ​​of the correlation strength of the neighboring order correlations in each composite service execution path; ② Select the combined service with the largest average association strength to obtain the best combined service execution path.

Images