Search

The industry has embraced digitalisation leading to a greater reliance on models derived from data. Understanding and getting insights into the models generated by machine learning algorithms is a challenge due to their non-explicit nature. Explainable artificial intelligence (XAI) is to enhance understanding of the digital models and confidence in the results they produce. The paper makes two contributions. First, the XRule algorithm proposed in the paper generates explicit rules meeting user's preferences. A user may control the nature of the rules generated by the XRule algorithm, e.g. degree of redundancy among the rules. Second, in analogy to federated learning, the concept of federated explainable artificial intelligence (fXAI) is proposed. Besides providing insights into the models built from data and explaining the predicted decisions, the fXAI provides additional value. The user-centric knowledge generated in support of fXAI may lead to discovery of previously unknown parameters and subsequently models that may benefit the non-explicit and explicit perspectives. The insights from fXAI could translate into new ways of modelling the phenomena of interest. A numerical example and three industrial applications illustrate the concepts presented in the paper. [ABSTRACT FROM AUTHOR]

Data has become a high-value commodity in manufacturing. There is a growing realisation that the data-driven applications could become strong differentiators of manufacturing enterprises. To guide the developments in digitisation, a widely accepted framework is needed. In the absence of the universal framework, the components making a digital enterprise are captured in an example framework that is introduced in the paper. The adoption of new technology and software solutions has increased complexity of manufacturing systems. In addition, new product introductions have become more frequent and the demand more variable. A digital space enables optimisation and simulation of decisions before their realisation in the physical space. Predictive modelling with its time dimension is a valuable actor in the digital space. Three challenges of predictive modelling such as model complexity, model interpretability, and model reuse are identified in this paper. The coverage of each challenge in the literature is illustrated with the recently published papers. The main aspects of these challenges and the synthesis of the developments in digital manufacturing are articulated in the form of eight observations that could guide the future research. [ABSTRACT FROM AUTHOR]

The resilience of a production system is determined by its capability to respond to internal breakdowns and/or external disruptions and recover. In conventional production systems, internal disruptions such as machine breakdown are handled by parallel stations and storage buffers, which come at a cost. In this paper, we propose to use reconfigurable machines (RMs) and movable buffers (MBs) to increase the resilience of a production system. The production system is modelled using a modified Markov chain model. To reduce the computational effort, an iterative method is adopted for the production lines that have many RMs and MBs. The resilience of the production system is evaluated by a combination of production loss, steady production rate with threshold, work-in-process in Idle-area of MBs, process time of work-in-process in Idle-area of MB with threshold, and investment return. Two production systems are analysed, one with 3 operations and the other with 10 operations. The computer simulation results indicate that the resilience of a production system can be improved by more than 9% by RMs and MBs. Finally, a set of guidelines for design production systems with RMs and MBs are also given. [ABSTRACT FROM AUTHOR]

Globalisation of the manufacturing and service industry has increased complexity of the flow of materials and goods, magnified dependency on the underlying network, and made the industry more vulnerable to the changing market conditions. As manufacturing and service industry undergo transformation, an opportunity to rethink the design of future enterprises has emerged. Six enablers of miniversal manufacturing are discussed: digitisation, open manufacturing, service orientation, shared manufacturing, sustainability, and resilience. These enablers will form properties of universal manufacturing, with adaptability and affinity that are emerging. A universal manufacturing enterprise will be formed based on the distributed manufacturing facilities. The emerging standards for interoperability of systems needed for universal enterprises are discussed. The data and modelling standards will enable the synthesis of digital models into universal enterprises. Though there is no global standard for the representation of digital manufacturing models in a cloud, the existing process modelling methodologies and languages may offer the solutions needed. The evolution of production systems is illustrated with three snapshots, dedicated manufacturing, distributed manufacturing, and universal manufacturing. The modelling approach followed in this paper is bottom-up rather than top-down followed in the literature on modern manufacturing. [ABSTRACT FROM AUTHOR]

The transformation of the manufacturing industry over the last two decades has been largely inspired by data. Digitisation has made its mark in different areas of manufacturing, from digital materials and processes to data science in decision-making. The digital manufacturing is evolving towards universal manufacturing that is highlighted in this paper. Different manufacturing initiatives are benchmarked and their relationship to universal manufacturing is demonstrated. The representation of enterprises in the universal manufacturing cloud is discussed. Product- and process-based specifications of digital enterprises are defined. Using these specifications, the enterprise configuration algorithm proposed in the paper is applied for the selection of component models. Two different representations, a node–node and an input–output matrix, of the digital component models are considered. The extended topological sorting algorithm is applied to construct an integrated digital model. [ABSTRACT FROM AUTHOR]

Open systems have been of interest to the research and industrial community for decades, e.g. software development, telecommunication, and innovation. The presence of open manufacturing enterprises in a cloud calls for broadly interpretable models. Though there is no global standard for representation of digital models of processes and systems in a cloud, the existing process modelling methodologies and languages are of interest to the manufacturing cloud. The models residing in the cloud need to be configured and reconfigured to meet different objectives, including complexity reduction and interpretability which coincide with the resilience requirements. Digitisation, greater openness, and growing service orientation of manufacturing offer opportunities to address resilience at the design rather than the operations stage. An algorithm is presented for complexity reduction of digital models. The complexity reduction algorithm decomposes complex structures and enhances interpretability and visibility of their components. The same algorithm and its variants could serve other known concepts supporting resilience such as modularity of products and processes as well as delayed product differentiation. The ideas introduced in the paper and the complexity reduction algorithm of digital models are illustrated with examples. Properties of the graph and matrix representations produced by the algorithm are discussed. [ABSTRACT FROM AUTHOR]

Manufacturing is undergoing transformation driven by the developments in process technology, information technology, and data science. A future manufacturing enterprise will be highly digital. This will create opportunities for machine learning algorithms to generate predictive models across the enterprise in the spirit of the digital twin concept. Convolutional and generative adversarial neural networks have received some attention of the manufacturing research community. Representative research and applications of the two machine learning concepts in manufacturing are presented. Advantages and limitations of each neural network are discussed. The paper might be helpful in identifying research gaps, inspire machine learning research in new manufacturing domains, contribute to the development of successful neural network architectures, and getting deeper insights into the manufacturing data. [ABSTRACT FROM AUTHOR]

Data have been utilised in production research in meaningful ways for decades. Recent years have offered data in larger volumes and improved quality collected from diverse sources. The state-of-the-art data research in production and the emerging methodologies are discussed. The review of the literature suggests that production research enabled by data has shifted from that based on analytical models to data-driven. Manufacturing and data envelopment analysis have been the most popular application areas of data-driven methodologies. The research published to date indicates that data mining is becoming a dominant methodology in production research. Future trends and opportunities for data-driven production research are presented. [ABSTRACT FROM AUTHOR]

Manufacturing has evolved and become more automated, computerised and complex. In this paper, the origin, current status and the future developments in manufacturing are disused. Smart manufacturing is an emerging form of production integrating manufacturing assets of today and tomorrow with sensors, computing platforms, communication technology, control, simulation, data intensive modelling and predictive engineering. It utilises the concepts of cyber-physical systems spearheaded by the internet of things, cloud computing, service-oriented computing, artificial intelligence and data science. Once implemented, these concepts and technologies would make smart manufacturing the hallmark of the next industrial revolution. The essence of smart manufacturing is captured in six pillars, manufacturing technology and processes, materials, data, predictive engineering, sustainability and resource sharing and networking. Material handling and supply chains have been an integral part of manufacturing. The anticipated developments in material handling and transportation and their integration with manufacturing driven by sustainability, shared services and service quality and are outlined. The future trends in smart manufacturing are captured in ten conjectures ranging from manufacturing digitisation and material-product-process phenomenon to enterprise dichotomy and standardisation. [ABSTRACT FROM AUTHOR]

Satisfying customer expectations is of paramount importance in today's markets. A customer expects to receive a product that meets as best as possible his/her expectations. To satisfy diversified requirements companies may focus on product families. For a specific customer the company has to select within a product family what will make a valid product. Then, in a just-in-time environment, the suppliers may have to provide the exact subassemblies corresponding to each product in a predefined time. This integration takes place by designing modules for the supplier. This paper proposes to extract customers' behaviour patterns, in terms of the components required, using data-mining and entropy maximisation when selecting modules to be manufactured. Different methods for selection of modules are proposed. Computational tests are performed to evaluate performance of the selection methods with respect to the specified assembly time/resource level. [ABSTRACT FROM AUTHOR]

Customers benefit from the ability to select their desired options to configure final products. Manufacturing companies, however, struggle with the dilemma of product diversity and manufacturing complexity. It is important, therefore, for them to capture correlations among the options provided to the customers. In this paper, a data mining approach is applied to manage product diversity and complexity. Rules are extracted from historical sales data and used to form sub-assemblies as well as product configurations. Methods for discovering frequently ordered product sub-assemblies and product configurations from 'if-then' rules are discussed separately. The development of the sub-assemblies and configurations allows for effective management of enterprise resources, contributes to the innovative design of new products, and streamlines manufacturing and supply chain processes. The ideas introduced in this paper are illustrated with examples and an industrial case study. [ABSTRACT FROM AUTHOR]

The assemble-to-order strategy delays the final assembly operations of a product until a customer order is received. The modules used in the final assembly operation result in a large product diversity. This production strategy reduces the customer waiting time for the product. As the lead-time is short, any product rework may violate the delivery time. Since quality tests can be performed on the stocked modules without impacting the assembly schedule, the quality of the final assembly operations should be the focus. The data-mining approach presented in this paper uses the production data to determine the sequence of assemblies that minimizes the risk of producing faulty products. The extracted knowledge plays an important role in sequencing modules and forming product families that minimize the cost of production faults. The concepts introduced in the paper are illustrated with numerical examples. [ABSTRACT FROM AUTHOR]

In this paper basic concepts of machine learning and data mining are introduced. Machine learning algorithms extract knowledge from diverse data bases that can be used to build decision-making systems. For example, based on the operational engineering data, equipment faults can be detected, the number of items to be ordered can be predicted, optimal control parameters can be determined. A framework for organizing and applying knowledge for decision-making in manufacturing and service applications is presented. The framework uses decision-making constructs such decision tables, decision maps, and atlases. It offers a new data-driven paradigm of importance to modern manufacturing and service organisations. Examples of data mining applications in industrial, medical, and pharmaceutical domains are presented. It is envisioned that the data-driven framework presented in the paper will enhance these applications. [ABSTRACT FROM AUTHOR]

This article introduces the topic of data mining which is the focus of this issue of the periodical. Data mining aids in the analysis of large databases and discovery of trends, patterns and knowledge. Many applications in engineering design, manufacturing and logistics engineering now incorporate data mining functions. This issue looks at data mining's theory and application as well as that of text mining, web mining, and image mining as it applies to engineering design, manufacturing and logistics engineering.

A successfully adopted and implemented enterprise resource planning (ERP) system does not automatically guarantee full benefits. It is important that ERP is operated as planned and thus provides the real-time information with a desired level of process integration. Any ERP system pushes a company toward full process integration and solves the fragmentation of information. However, the tight process integration involves operational issues that must be carefully managed. Thus, a conjecture can be made that a centrally coordinated support system is required to assist ERP users and administrators find problems, perform tedious validation and verification, and maintain the process integration of ERP with great consistency. This paper proposes an ERP operations support system (EOSS) that aims to achieve and maintain the process integration of ERP. EOSS monitors the operational status of process integration, prevents anomalies as early as possible and conducts repairs as soon as possible. [ABSTRACT FROM AUTHOR]

Companies design and manufacture widely diversified products to satisfy the needs of their customers and markets. Two issues important to achieving this aim are discussed. The first concerns adequate diversity for a particular market. The second concerns the management and manufacture of products within an acceptable lead time and an acceptable cost. The two issues are examined with a methodology for the design of products families. This methodology is based on a data-mining approach and it focuses on the analysis of functional requirements. [ABSTRACT FROM AUTHOR]

Despite the efforts in developing Petri net models for manufacturing control and scheduling, the generation of Petri net models cannot be automated for agile manufacturing control and scheduling without difficulties. The problems lie in the complexity of Petri net models. First of all, it is difficult to visualize the basic manufacturing process flow in a complex Petri net model even for a Petri net modelling expert. The second problem is related to the complexity of using Petri net models for manufacturing system scheduling. In this paper, a decomposition methodology in automatic generation of Petri nets for manufacturing system control and scheduling is developed. The decomposition methodology includes representing a manufacturing process with the Integrated Definition 3 (IDEF3) methodology, decomposing the manufacturing process based on the similarity of resources, transforming the IDEF3 model into a Petri net control model, and aggregating sub Petri net models. Specifically, a sequential cluster identification algorithm is developed to decompose a manufacturing system represented as an IDEF3 model. The methodology is illustrated with a flexible disassembly cell example. The computational experience shows that the methodology developed in this paper reduces the computational time complexity of the scheduling problem without significantly affecting the solution quality obtained by a simulated annealing scheduling algorithm. The advantages of the methodology developed in this paper include the combined benefits of simplicity of the IDEF3 representation of manufacturing processes and analytical and control properties of Petri net models. The IDEF3 representation of a manufacturing process enhances the manmachine interface. [ABSTRACT FROM AUTHOR]

This paper reviews some of the current developments in computer-aided fixture design (CAFXD), and proposes directions for future research initiatives. The literature has indicated a growing interest in the development of computer-aided fixture design systems. The motivation for this manufacturing research is stimulated by the desire to reduce setup and production costs, and the potential rewards of automation. Most of the papers published have centred on the micro aspects of fixture design that addressed singular problems; but as the research matures, more systems will be directed towards an integration with other computer-aided engineering tools for total fixture design. [ABSTRACT FROM AUTHOR]

Concurrent design should result in reduction of the duration of a design project, cost reduction, and better quality of the final design; however, it may increase the complexity of the design process and make it more difficult to manage. In this paper, an algorithm is developed for organizing design activities in order to effectively produce an acceptable design. The relationship among design activities is represented with an incidence matrix and the corresponding directed graph. The design process is simplified by identifying and analysing design activities that are coupled. The algorithm presented in the paper generates a sequence of design activities such that the number of cycles is minimized, i.e. the product development time is reduced. The concepts presented are illustrated with examples. [ABSTRACT FROM AUTHOR]

This paper discusses two formulations of the group technology (GT) problem. The first one involves basic and alternative process plans without bottleneck parts. The second formulation considers basic process plans and bottleneck parts or bottleneck machines. The algorithms developed in the paper are illustrated with numerical examples. Computational results are also provided. [ABSTRACT FROM AUTHOR]

In this paper, design specifications are represented with an AND/OR tree. The logical tree allows for representation of alternative solutions in the process of formulating design specifications. Various approaches for search of the logical tree are discussed. A generalized search that is based on a set of production rules is used for decomposition of requirements and functions as well as matching the requirements with functions. [ABSTRACT FROM AUTHOR]

Conceptual design is a process of converting design requirements into an acceptable design solution. An object-oriented system for conceptual design is presented. The process of component synthesis is guided by production rules. An object-oriented programming paradigm is chosen since it simulates the designer's way of thinking during conceptual design. The system is implemented in Smalltalk-80 programming language and environment. Besides providing all the characteristics of an object-oriented language, Smalltalk-80 allows for development of highly user-friendly interfaces. The design synthesis process is presented using a graphical and textual description. [ABSTRACT FROM AUTHOR]

Design of complex products or large scale systems involves hundreds of resources (designers, analysts, computers, software systems, and procedures) and thousands of design activities. In this paper, a methodology is presented for decomposition of the design task into activities and modules. The methodology is based or clustering of design activities into groups that allow effective organization on resources required in the design process. A knowledge-based approach is used for managing design activities. The system performs analysis aiming at exploring concurrency and reducing the design project makespan. [ABSTRACT FROM AUTHOR]

In this paper the machine layout problem in an automated manufacturing system is addressed. Two models for solving the machine layout problem are developed. A knowledge-based system for machine layout (KBML) is presented. KBML combines the optimization and expert system approaches and considers quantitative as well as qualitative factors while solving the machine layout problem. The system is coded in Common LISP and implemented on a Symbolics 3650 machine. It is illustrated with a numerical example. [ABSTRACT FROM AUTHOR]

In this paper a knowledge based system (EXGT-S.) for solving the generalized group technology problem is presented. The formulation of the group technology problem involves constraints related to machine capacity, material handling system capabilities, machine cell dimensions and technological requirements. It has been developed for all automated manufacturing system. EXGT-S is based on the tandem system architecture presented in Kusiak (1987). It considers alternative process plans and multiple machines. EXGT-S takes advantage of the developments in expert systems and optimization. Two basic components of the knowledge based system, namely the expert system and heuristic clustering algorithm are discussed. Each partial solution generated by the clustering algorithm is evaluated by the expert system which modifies search directions of the algorithm. [ABSTRACT FROM AUTHOR]

Focuses on the matrix and integer programming classes of clustering models. Relationship of the matrix model with the p-median model and the classical group technology concept; Impact of the models on the quality of process families and machine cells.

Discusses issues related to flexible manufacturing systems (FMS). Application of technology in manufacturing; Subsystems of an FMS; Trends in FMS development.

Presents a unified pattern recognition approach to the representation of part geometry. Relationships among types of geometry modelling; Matrix representation of part geometry; Operations on matrix representation.

Kanban systems have been successfully used to reduce inventory and improve customer service in many industries. Their advantages and limitations are discussed in this paper. Although numerous models have been developed to overcome these limitations, the studies published did not consider some of the very essential characteristics of manufacturing systems and the value added to parts when the pull or push principle was applied. This paper presents a push-pull approach which pushes through certain manufacturing stages and pulls elsewhere based on the characteristics and value added at these stages. A generalized labelcorrecting algorithm to determine the desired stages of an IDEF3 model is developed and an industrial case study is presented. This methodology is suited for a repetitive manufacturing environment of a mixed type, i.e. job shop and flow shop, and leads to achieving the following objectives: (1) reduced in-process inventory; (2) shortened delivery lead times while maintaining or improving delivery dependability; (3) increased productivity and machine utilization. [ABSTRACT FROM AUTHOR]

The concept of agile manufacturing is driven by the need to quickly respond to the changing customer requirements. Agile manufacturing demands a manufacturing system to be able to produce efficiently a large variety of products and be reconfigurable to accommodate changes in the product mix and product designs. The manufacturing system reconfigurability and product variety are critical in agile manufacturing. The concept of agility has an impact on design of assemblies. To implement agile manufacturing, methodologies of design for agile manufacturing are needed. Design for agile assembly is accomplished by considering operational issues of assembly systems at the early product design stage. In this paper, three rules applicable to the design of products for agile assembly from an operational perspective are proposed. These rules are intended to support the design of products to meet the requirements of agile manufacturing. Illustrative examples are provided to demonstrate the potential of the design rules. Procedures and algorithms for implementing these rules are presented. [ABSTRACT FROM AUTHOR]

The product development process is subject to randomness of duration of design activities and a path followed through the network of activities. In this paper, the qualitative relationships between various attributes of the design process and the corresponding design process variables are captured using the house of quality and then transformed into quantitative relationships. The impact of the process control variables on the design process attributes is discussed. The problem of determining optimal values of the design process variables to maximize the combined quality index of the critical design activities is modelled as a geometric programming problem. [ABSTRACT FROM AUTHOR]