Knowledgeable Manufacturing System Research Articles

Matching decision method for knowledgeable manufacturing system and its production environment

To secure the quick response of knowledgeable manufacturing system (KMS) to the dynamic production environment and its desirable adaptability and competitiveness, we propose a matching decision method that is based on the improved support vector machine (ISVM for short), and the production environment. Taking into account the uncertainty and fuzziness of the production environment, the triangular fuzzy numbers are introduced to represent the uncertain input factors. Independent penalty coefficients are employed for different categories to address the problem of unbalanced samples. To meet the requirement for classifying small, uncertain input, and unbalanced samples, an improved SVM model based on triangular fuzzy theory is put forward. Considering the mutagenic factor and dynamic weight, we improve the particle swarm algorithm to optimize the model parameters. The matching categories of KMS and dynamic production environment are defined, and the corresponding matching decision method based on ISVM model is built. Case study shows that the proposed ISVM matching decision method is feasible and effective.

A scheduling procedure for a flow shop–like knowledgeable manufacturing cell with self-evolutionary features

As a component of knowledgeable manufacturing systems, the structure of flow shop–like knowledgeable manufacturing cells is similar to that of a flow shop, thus representing an NP-hard issue. Here, we propose a self-evolutionary algorithm that exhibits learning ability and is composed of learning and scheduling modules. Unlike traditional scheduling algorithms, whose performances remain unchanged when the procedure is coded, the performance of the algorithm proposed in this study gradually improves as the learning process continues. The self-evolutionary ability is realized through the training of a hybrid kernel support vector machine. The hybrid kernel support vector machine was designed to approximate the value of the Q-function to select the appropriate action for the scheduling module and thus to obtain the optimal solution. An iterative process of value based on the Q-learning was adopted to train the hybrid kernel support vector machine to gradually enhance the algorithm’s efficiency and accuracy. The extracted state features of the flow shop–like knowledgeable manufacturing cells serve as inputs to hybrid kernel support vector machine for easy generalization of the learning results. The action exerted on a feasible solution is also defined as the input of the hybrid kernel support vector machine. The computational results show that the performance of the proposed procedure improves as the learning process progresses. Data from the computation and comparisons with other algorithms verify the validity and efficiency of the proposed algorithm.

The retrieval method of KMS knowledge meshes by complexity analysis

Knowledgeable manufacturing system (KMS) transforms all types of advanced manufacturing modes into corresponding knowledge meshes (KMs) and selects the best combination of KMs to satisfy enterprise requirements. Efficiently retrieving and reconfiguring KMs can reduce the complexity of new KMs gained by self-reconfiguration operations and enhance its practicability. This paper presents the method for measuring the KM complexity based on entropy and that for fuzzily classifying and retrieving for KMs based on granularity. Utilising the intrinsic information of KM, knowledge capacity function based on entropy is introduced to measure the KM complexity, and proved to be a monotone function of the number, measure, coefficient and weight of elements in KM. Properties of the KM operations are kept. Taking quality, quantity and complexity into account, the similarity function is defined. As revealed by our analysis, this function is of similarity both in the sense of matching, and in the mode of gaining KM. Then, the KMs in the KM base are fuzzily clustered. The number of classes is not fixed in advance, but can be dynamically adjusted. Each clustering centre is the best state corresponding to certain demands and has the minimum complexity degree. KM features are quantised in importance using the weight vectors. The search space is determined by centring at the clustering, which converts the problem from fine-grained space to coarse-grained space. Our tests and software package developed have proved the method to be quite effective.

Self-Evolution of an Assembly Workshop in Knowledgeable Manufacturing Environment

In this study, the self-evolution problem of knowledgeable manufacturing systems is studied by taking an assembly workshop as an example. The rolling horizon procedure (RHP) is adopted to implement the self-evolution process of the workshop. The whole dynamic self-evolution process is decomposed into several static decision processes. At each decision point, a static decision sub-problem needs to be solved. A general mathematical model of these sub-problems is built, and a bi-level genetic algorithm (BiGA) is designed. Simulation results show that the model and algorithm are feasible and effective. By comparison, the system with self-evolution operations has a better production performance.

Clustering and selection of knowledge meshes of knowledgeable manufacturing systems based on decomposition of fuzzy relational data

This article presents a study on the clustering and selection of knowledge meshes in the knowledgeable manufacturing system that transforms all types of advanced manufacturing modes into corresponding knowledge meshes and selects the best combination of knowledge meshes to satisfy enterprise requirements. The appropriate knowledge mesh for enterprises not only includes the matching of knowledge mesh function but also that of performance perfection and structure. Thus, the similarity degree of knowledge meshes, whose properties are proved to relate to the operations of knowledge meshes, is constructed from the functional matching, the perfection degree and the layer number of lowest-layer knowledge point. The similarity values, taken as cluster data, are used to construct the fuzzy relational matrix to compress the high-dimensional feature space. The decomposition of the matrix is transformed into an optimization problem solved by the gradient method. The knowledge meshes with higher membership degree in each class are taken as reference knowledge meshes to identify user’s requirements exactly. The comparison of target knowledge mesh with reference knowledge meshes definitely narrows down the knowledge mesh selection to a certain type. Based on the above, the knowledge mesh clustering and selection method is exemplified. The results show that the proposed method works well in narrowing the search range and clarifying user requirements.

An interoperable adaptive scheduling strategy for knowledgeable manufacturing based on SMGWQ-learning

To address the uncertainty of production environment in knowledgeable manufacturing system, an interoperable knowledgeable dynamic-scheduling system based on multi-agent is designed, wherein an adaptive scheduling mechanism based on the state membership grade weighted Q-learning (known as SMGWQ-learning) is proposed for guiding the equipment agent to select proper scheduling strategy in a dynamic environment. To avoid the side effect of large state space and minimize errors between the clustering and real states, the state membership grade, defined as weight coefficients, is incorporated into the weighted Q-value update so that several Q-values can be updated simultaneously in an iteration. Results from our convergence analysis and simulation experiments show the effectiveness of the proposed strategy that endows the scheduling system with higher intelligence, interoperability and adaptability to environmental changes by self-learning.

The Structure Principle of Knowledge Mesh and its Application

Aiming at the rationality of new knowledge mesh, which is obtained from the self-reconfiguration of knowledgeable manufacturing systems, the inherent structure principle of the knowledge mesh is studied. Firstly, the sets of the knowledge points are classified by the equivalent relations, and the function topological space is constructed. The relationships between the sets of the knowledge points and the function topological space are established by the real set. In the function topological space, the conclusions of the base are given, which show the essential mapping relationships of the knowledge mesh, the topological space and the base. After knowing the principle, the structure of the knowledge mesh is reduced by using the equivalent relations, the base and the basic knowledge points. The completeness of the complex relationships in the knowledge mesh is studied by the transitive closure. These conclusions can reduce the structure and enhance the rationality of the knowledge mesh. Finally the given example shows that these theories are effective.

Automatic construction and optimization of knowledge mesh for self-reconfiguration of knowledgeable manufacturing system

This paper deals with an approach to the automatic construction and optimization of the knowledge mesh (KM) based on the user’s function requirements. Once a KM multiple set operation expression is obtained, a new KM can be inferred from the expression by the developed KM-based inference engine and transformed into its corresponding KMS (knowledgeable manufacturing system) software automatically by the developed automatic program construction software so as to realize the self-reconfiguration of the KMS. Thus, the automatic construction and optimization of a KM multiple set operation expression is equivalent to the automatic construction and optimization of its corresponding KM and KMS software. To explore the automatic construction and optimization of the new KM by the user’s function requirements, an automatic construction procedure of a KM aiming at the user’s maximum function–satisfaction is proposed. Firstly, the fuzzy function–satisfaction degree relationships of the users’ requirements for the KM functions are defined, and so are the multiple fuzzy function–satisfaction degrees of the relationships. Secondly, operations (union, intersection and minus) on both fuzzy and multiple fuzzy function–satisfaction degrees are proposed and clarified, along with the proof that there exists a one-to-one mapping between the KM multiple set operation expression and the KM-function–satisfaction degree expression. Then, the optimization model of the KM multiple set operation expression is constructed and proved to be very NP-hard. And finally, the KM multiple set operation expression is optimized by the hybrid genetic-tabu algorithm, with the steps of the KM’s automatic construction presented in detail as well. Based upon the above, the KM’s automatic construction and optimization are illustrated by an actual KM example which corresponds to the management information system (MIS) software used in a vehicle body plant. The proposed approach proves to be very effective.

Competitive diffusion process of repurchased products in knowledgeable manufacturing

This paper presents a diffusion model to explain the competitive diffusion of the repurchased products in knowledgeable manufacturing. The acute market competition accelerates the products’ improvement, which requires that the manufacturing enterprises be highly capable of rapid reaction by means of knowledgeable manufacturing. To forecast the diffusion behavior effectively enables the realization of knowledgeable manufacturing system (KMS) which targets T (time), Q (quality), C (cost), S (service), and E (environment). Various diffusion models have emerged since Bass model was firstly proposed in 1969. A nonlinear model of the repurchased competitive products is proposed on the basis of the product diffusion analysis. By taking the frequently purchased products as example, the stability of the model is examined in light of the qualitative theory of differential equations and proved by the approximate linearization method. As the qualitative analysis reveals, between the two frequently purchased products competing in the same market, one undoubtedly occupies a fixed market share while the other may finally be eliminated from the market. A special case of the problem is that both products are one-time-purchased. With the corresponding model given, the qualitative analysis shows that either of the products occupies a market share, the size of which is determined by the product’s competitive strength and the new product’s time-to-market. A system dynamics model is then established and simulated by vensim. The result is consistent with that of the qualitative analysis.

A new bottleneck detecting approach to productivity improvement of knowledgeable manufacturing system

Manufacturing systems are usually restricted by one or more bottlenecks. Identification of bottleneck is a key factor to improving the throughput of a production system. However, locating the bottleneck is no easy task. This paper proposes a novel method of bottleneck detecting for knowledgeable manufacturing system (KMS). Presented a net-like model of the knowledgeable manufacturing system for bottleneck detection, which adapts well to the bottleneck analysis in flexible product lines. Based on the model, the concept of entire production net is defined and an approach to identifying bottlenecks in entire production nets is developed and proven effective theoretically. A self-learning method is introduced for storing the knowledge of bottlenecks and their conditions in the knowledge base to detect which cells are required to upgrade their capacity. Validity of the approach is verified by the numerical experiment.

A deadlock avoidance strategy for knowledgeable manufacturing systems

Deadlock is a highly unfavourable situation in manufacturing systems, and the development of a deadlock-avoidance policy for resource allocation has received significant attention. Based on finite automata theory, the concept of a distributed deadlock supervisor is proposed to avoid deadlock in knowledgeable manufacturing cell. The proposed approach is insensitive to system scale and improves the efficiency of the modelling and operation of automata. The auto-correlation operation is presented, and the knowledgeable manufacturing cell is divided into a number of submodules, and the corresponding distributed deadlock supervisors are constructed. A state backtrack algorithm is given to ensure full reachability of deadlock supervisor, and thus a deadlock-free strategy for the knowledgeable manufacturing cell is obtained. This can be used to monitor the status of the knowledge manufacturing cell in real-time and guarantees deadlock-free running of the system. A case study is presented to demonstrate the feasibility and effectiveness of the proposed strategy.

Control of knowledgeable manufacturing cell with an unreliable agent

Aiming at the task control problems existing in the knowledgeable manufacturing system, the concept of state jump system is proposed to analyze a knowledgeable manufacturing cell with an unreliable agent. The uncertain factors of the knowledgeable manufacturing cell are addressed in the task control model by utilizing a self-learning method of probability distribution parameters of stochastic events. With the state jump system given, the task control problem is greatly simplified that the optimal task control strategy of the manufacturing cell can be obtained by the combination of the uniform technology and the stochastic dynamic programming. The objective function can be stabilized to a certain extent for different initial conditions, which verifies the feasibility of the control strategy. Compared to the random control and maximum control principles, the objective function value of the optimal control strategy in this paper is relatively low, which confirms the validity of the control strategy.

Decision-making in self-reconfiguration of a knowledgeable manufacturing system

This paper presents the two decision-making methods for reconfiguring knowledge meshes (KMs) and selecting the best from several reconfigured KMs in the self-reconfiguration of a knowledgeable manufacturing system (KMS). Based on the fuzzy requirement relationships for KMs, the operations on fuzzy requirement degrees for KMs are defined and the requirement evaluation method for the KM is proposed together with its decision-making method, which is used for the KM reconfiguration. Next, a grey decision-making tree method is proposed to select the best from several reconfigured KMs. First, the factors influencing the decision-making on the selection of reconfigured KMs are analysed. Then, the conceptual model of decision-making factors is given, and the factors are evaluated by various methods such as investigation, evaluation, and functional analysis. On the basis of relational analysis, the grey decision-making tree model is proposed, and the detailed decision-making steps are provided. Finally, the KM reconfiguration and the reconfigured-KM selection are exemplified. The decision-making result fully indicates that the methods can effectively solve the decision-making problem in the KMS self-reconfiguration.

A new complicated-knowledge representation approach based on knowledge meshes

This paper presents a new complicated-knowledge representation method for the self-reconfiguration of complex systems such as complex software systems, complex manufacturing systems, and knowledgeable manufacturing systems. Herein, new concepts of a knowledge mesh (KM) and an agent mesh (AM) are proposed along with a new KM-based approach to complicated-knowledge representation. KM is the representation of such complicated macroknowledge as an advanced manufacturing mode, focusing on knowledge about the structure, functions, and information flows of an advanced manufacturing system. The multiple set, KM, and the mapping relationships between both, are then formally defined. The union, intersection, and minus operations on the multiple sets are proposed, and their properties proved. Then, the perfectness of a KM, the redundancy set between the two KMs, and the multiple redundancy set on the redundancy set are defined. Three examples are provided to illustrate the concepts of the KM, multiple set, multiple redundancy set, and logical operations. On the basis of the above, the KM-based inference engine is presented. In logical operations on KMs, each KM is taken as an operand. A new KM obtained by operations on KM multiple sets can be mapped into an AM for automatic reconfiguration of complex software systems. Finally, the combination of two real management modes is exemplified for the effective application of the new KM-based method to the self-reconfiguration of complex systems. It is worth mentioning that KM multiple sets can also be taken as a new formal representation of software systems if their corresponding AMs are the real software systems.

NEW APPROACH TO KNOWLEDGE REPRESENTATION FOR ADVANCED MANUFACTURING MODES

A new approach to knowledge representation for advanced manufacturing modes based on knowledge mesh (KM) and agent mesh (AM) is proposed to solve the self-reconfiguration problem of knowledgeable manufacturing systems. First, the definition of KMs is given and the relationships between KMs and AMs are presented. The multiple set and its mapping relationships onto the KM are then defined. The union, intersection and minus operations on the multiple sets of KMs are also defined and their related properties are presented. Physical meanings of properties of operations on the multiple sets are studied. Properties of operations on the multiple sets are compared with those of operations on the classical sets. Then the perfectness of a KM, the cut set between the two KMs, and the multiple cut set on the cut set are defined. On the basis of those above, the relational-database-based realization and the inference engine of KMs are studied. The self-reconfiguration methods based on both KMs and AMs are presented through two examples of self-reconfiguration of knowledgeable manufacturing systems, which verifies the validity of these methods.

A study on self-reconfiguration of a knowledgeable manufacturing system

This paper deals with a new approach to the self-reconfiguration of the knowledgeable manufacturing system (KMS), which is based on knowledge meshes and agent meshes. To this end, new concepts of a knowledge mesh (K-M) and an agent mesh (A-M) are proposed as well as a new K-M and A-M based approach to knowledge representation for advanced manufacturing modes. A multiple set on the K-M is also formally defined as well as the mapping relationships between the multiple sets and the K-M. The union, intersection and minus operations on the multiple sets and the algorithms of K-M operations are proposed. In addition, preprocessing rules and the mapping between the multiple sets and the new K-M are explored in detail. On the basis of these, the relational database-based realizations of the K-M and its multiple sets are presented. Functional structure and realization of the self-reconfiguration subsystem of KMS are also presented. Finally, the self-reconfiguration method based on the K-M and A-M is exemplified by the self-reconfiguration of the management information system (MIS) software used in a vehicle body plant at Nanjing in China, which shows the method to be very effective.