Adaptive Manufacturing Systems Research Articles

Cognitive work in future manufacturing systems: Human-centered AI for joint work with models

Manufacturers perpetually adapt their systems to meet unforeseen events, new objectives, competition, and improved understanding of processes. In that human-directed work, models mediate an enduring relationship between production resources and engineers. Accommodating new understanding in the models controlling production can lead to more effective manufacturing. That work has previously been the province of quality programs such as Six Sigma, but is now fertile ground to study human-computer interaction about that enduring relationship mediated by models. Can AI augment human capability in the arcane work of formulating and refining models? This question is relevant to complex system engineering generally, not just manufacturing. In answering this question, this paper adapts Klein’s flexecution for use in adaptable manufacturing systems. Theory flexecution, the methodical refinement of models, points to human-computer interactions that emphasize the roles of models, explanation, and machine agents that recognize the engineer’s goals. This perspective article illustrates these ideas with an example of formulating models for production scheduling.

Adaptive edge finishing process on distorted features through robot-assisted computer vision

Robotic deburring has emerged as a transformative solution in finishing processes, offering independence from operator influence to enhance product quality and reduce production costs. To do this, contour tracking is key to ensuring cutting tool contact and obtaining chamfers within narrow tolerances, especially on high added-value parts such as aeronautical turbomachinery. The persistent challenge involves accommodating workpiece variability, including deformations and burrs. Current probe positioning methods lack insights into burr morphology, while optical systems often require expensive scanners, resulting in time-intensive point cloud generation. Therefore, this paper presents an innovative industrial computer vision approach for adaptive deburring of functional edges. By processing the detected contour deviation signal, the proposed method generates tailored toolpaths based on real-time part conditions. The primary contribution lies in its ability to swiftly detect edges and burrs, dynamically adjusting machining toolpaths and conditions based on their dimensions, all through the rapid analysis of a simple image. This approach optimizes deburring operations, directing the cutting tool precisely where it is needed. Consequently, the default NC programming was enhanced with two new sub-programs tailored for deburring with conical tools—one for primary burr removal and another for uniform chamfering of distorted edges. The developed methodology can be easily applied in various industrial environments, offering a rapid and efficient means to improve chamfering of workpieces with variable contours. This research not only presents a practical solution for robotic deburring but also contributes to the advancement of adaptive manufacturing systems.

A Comprehensive Digital Model Approach for Adaptive Manufacturing Systems

In the current global economy, where rapid changes and constantly shifting market demands define the competitive landscape, adaptive manufacturing systems become essential for businesses striving to remain relevant and efficient. In the context of this growing need, this study focuses on planning as a part of adaptive manufacturing system. This methodology provides a systematic framework that spans from foundational groundwork to meticulous verification and validation phases. By employing advanced simulation techniques, seamless data integration, and process optimization, this methodology ensures the smooth realization of robust and efficient adaptive manufacturing systems. A detailed case study on competency islands showcases the versatility of this approach, demonstrating its efficacy in enhancing manufacturing agility and overall performance. As a significant contribution to the field of smart manufacturing, this methodology offers a structured blueprint for the realization of adaptive manufacturing systems.

PLC orchestration automation to enhance human–machine integration in adaptive manufacturing systems

Current approaches to manufacturing must evolve to respond to increasing demands for short product life cycles and customised products. Adaptive manufacturing systems integrate advanced technologies, automation, and data-driven methodologies to develop adaptable, efficient, and responsive production processes. Central to this concept is the emphasis on human involvement and fostering synergy between human operators and the manufacturing system. Significant changes to the system’s controller are required to achieve adaptivity, with programmable logic controllers (PLCs) being a common controller type. After the necessary changes to the configuration of the manufacturing system, the PLC should be reconfigured to orchestrate the new required behaviour. Automated reconfiguration is vital to rapidly responding to change, but some changes cannot be entirely achieved without human input in collaboration with automated methods. Conventional practices in PLC programming include manual, repetitive coding practices subject to errors. As a result, to ensure operational safety, the changes must be tested before being deployed to operations, ensuring it is error-free. This paper presents a methodology to automatically reconfigure the simulation environment and controller in response to a new product request. We automate the PLC code generation and testing practices to support and free up the operators when performing repetitive manufacturing reconfiguration tasks. The methodology is based on human learning, software automation, customised program development, knowledge graphs, and Graph Neural Networks (GNNs). The presented solution is a generic, vendor-agnostic, and interoperable solution that facilitates information exchange among multiple heterogeneous environments. Lastly, we have validated the methodology as a proof of concept at an adaptive assembly cell at the University of Nottingham in the United Kingdom.

Flexibility management and decision making in cyber-physical systems utilizing digital lean principles with Brain-inspired computing pattern recognition in Industry 4.0

ABSTRACT Industry 4.0 and its accompanying Cyber-Physical Manufacturing Systems in digitized have recently made new approaches to optimising production operations in manufacturing. The objective of this article is to evaluate how digital lean principles can support Industry 4.0 in pursuit of reducing non-value-added tasks from production processes, flexibility management and decision-making process with machine learning techniques. This research study comes under three levels, such as data level, information level, and knowledge level. Initially, the data from the CNC machine are collected via a Wireless Sensor Network (WSN). The collected data are stored in the cloud platform where the unwanted wastes of 7 Muda are removed by using the Value Stream Mapping (VSM 4.0) tool. It is critical in the big data world to have a systematic method for gathering, managing, and analysing data to gain valuable insights from it. Predictive maintenance provides a detailed examination of the detection, location and diagnosis of faults in related types of machinery. Accordingly, Big Data Analytics empowers the lean principle technique of Total Productive Maintenance (TPM) is suggested for avoiding potential failures to predict maintenance by allowing KPIs to be calculated in real-time. This approach requires high-performance procedures and adaptable manufacturing systems in the current digitalized lean. The information is then processed in the knowledge layer, utilizing the algorithm, rule and lean knowledge bases. The flexibility of manufacturing firms is determined by the adaptability of their shop floor processes. To meet these requirements the article developed pull control techniques of capacity slack CONWIP (CSC) control in digital lean production systems to guide the CPS deployment to offer flexible production systems. Reliable software systems are hoped to facilitate data analysis and autonomous decision-making. Finally, in the decision-making process, the article proposed the Brain-Inspired Computing of Structural and Syntactic (BIC-SS) pattern recognition method. The performance analysis of these findings is simulated in MATLAB software. Simulation can be done with the identification of ideal Kanban parameters like cycle time, lead time, delivery frequency and lot size. Furthermore, lean manufacturing improves company quality and productivity by decreasing waste and production costs, as well as adapting well to the many innovative systems that encourage the culture of change and quality inside organizations.

Towards adaptive digital twins architecture

The use of Digital Twins (DTs) for continuously optimising manufacturing systems under a constant stream of changes, also known as ”online optimisation”, is taken for granted by many authors but rarely demonstrated possible given the challenges in keeping a DT synchronised with its real system whilst using it to run look-ahead simulations. This research addresses this gap by demonstrating that online optimisation is achievable alongside real-time look-ahead simulation in DTs, even under constant changes in the system being modelled. The main enabling factor is a proposed architecture which can underpin a Digital Twin with Adaptive capabilities, or Adaptive Digital Twin (ADT). The capabilities include Real-time Simulation, Online Optimisation, and Adaptivity (RSO2A). The proposed ADT architecture is suitable for constantly changing production environments with unpredictable demands, for example, those envisioned to deliver the concept of mass personalisation, allowing customers to co-create and co-design products based on personal preferences. To demonstrate and validate the support of the ADT architecture for RSO2A, an Adaptive Manufacturing System (AMS) for mass personalisation is developed in silico. The AMS is underpinned by the proposed ADT architecture and simulated its operation and adaptation using realistic shoe personalisation scenarios. The simulation output demonstrates how the proposed architecture and the ADT built with it enable the AMS to maintain continuous production of personalised shoes and continuously re-configure its layout to adapt to new changes in the production environment.

Adaptive Cognitive Manufacturing System (ACMS) – a new paradigm

Innovation and transformative changes in products, manufacturing technologies, business strategies, and manufacturing paradigms have profoundly changed the manufacturing systems. In addition to being environmentally, economically socially sustainable, manufacturing systems are increasingly using intelligent technologies to be even more resilient, responsive, and adaptable. A new Adaptive Cognitive Manufacturing Systems (ACMS) paradigm, its drivers, enablers, and characteristics, including cognitive adaptation, is presented. Classification and definitions of four types of adaptability in manufacturing systems are included. Human-centric collaboration of workers and intelligent machines and applications, and the future of work in cognitive adaptive manufacturing systems are outlined. Cognitive Digital Twins (CDT), their features, evolution, and their use to support humans in intelligent, collaborative manufacturing settings are discussed. Industrial applications and case studies are used to illustrate the presented concepts and paradigms. Challenges and future research directions to achieve the ACMS paradigm and implement more intelligent, more adaptive, and sustainable manufacturing systems are presented. The presented novel concepts and technologies make significant contributions to the fast-evolving field of manufacturing systems. This pioneering research sheds light on many important future research topics and provides a road map and motivation for researchers in this field.

Knowledge Integration in Smart Factories

Knowledge integration is well explained by the human–organization–technology (HOT) approach known from knowledge management. This approach contains the horizontal and vertical interaction and communication between employees, human-to-machine, but also machine-to-machine. Different organizational structures and processes are supported with the help of appropriate technologies and suitable data processing and integration techniques. In a Smart Factory, manufacturing systems act largely autonomously on the basis of continuously collected data. The technical design concerns the networking of machines, their connectivity and the interaction between human and machine as well as machine-to-machine. Within a Smart Factory, machines can be considered as intelligent manufacturing systems. Such manufacturing systems can autonomously adapt to events through the ability to intelligently analyze data and act as adaptive manufacturing systems that consider changes in production, the supply chain and customer requirements. Inter-connected physical devices, sensors, actuators, and controllers form the building block of the Smart Factory, which is called the Internet of Things (IoT). IoT uses different data processing solutions, such as cloud computing, fog computing, or edge computing, to fuse and process data. This is accomplished in an integrated and cross-device manner.

Life Cycle Engineering 4.0: A Proposal to Conceive Manufacturing Systems for Industry 4.0 Centred on the Human Factor (DfHFinI4.0)

Engineering 4.0 environments are characterised by the digitisation, virtualisation, and connectivity of products, processes, and facilities composed of reconfigurable and adaptive socio-technical cyber-physical manufacturing systems (SCMS), in which Operator 4.0 works in real time in VUCA (volatile, uncertain, complex and ambiguous) contexts and markets. This situation gives rise to the interest in developing a framework for the conception of SCMS that allows the integration of the human factor, management, training, and development of the competencies of Operator 4.0 as fundamental aspects of the aforementioned system. The present paper is focused on answering how to conceive the adaptive manufacturing systems of Industry 4.0 through the operation, growth, and development of human talent in VUCA contexts. With this objective, exploratory research is carried, out whose contribution is specified in a framework called Design for the Human Factor in Industry 4.0 (DfHFinI4.0). From among the conceptual frameworks employed therein, the connectivist paradigm, Ashby’s law of requisite variety and Vigotsky’s activity theory are taken into consideration, in order to enable the affective-cognitive and timeless integration of the human factor within the SCMS. DfHFinI4.0 can be integrated into the life cycle engineering of the enterprise reference architectures, thereby obtaining manufacturing systems for Industry 4.0 focused on the human factor. The suggested framework is illustrated as a case study for the Purdue Enterprise Reference Architecture (PERA) methodology, which transforms it into PERA 4.0.

Roadmap to an adaptive assembly line for e-axles

Achieving high quality, high variety batch size production can be quite expensive. In this vision paper, the methodology of achieving this at low costs and the available technologies in the field of e-mobility production are described. The focus of this research lies in high adaptive and cognitive aspects in the assembly along with qualitative aspects. To match the high flexibility of a flexible manufacturing system while considering quantitative efforts, a use case of an e-axle assembly is being done. E-axle is chosen due to the ongoing electrification of mobility. Hence, a solution for implementing a set of methodologies for an adaptive manufacturing system with respect to assembly, quality and implementation efforts is shown. A Level of Practical Application matrix is presented of all the possible adaptive technologies that are feasible to implement in the e-assembly line.

(Self-)adaptiveness for manufacturing systems: challenges and approaches

Shrinking lot sizes and growing product variability demand frequent changes in manufacturing systems. Common manufacturing systems, however, are built for large, invariant lot sizes. (Self-)adaptive manufacturing systems can, on the other hand, react quickly to changes in both function and capabilities. This ability makes them suitable to meet new customer or production requirements. Introducing (self-)adaptiveness in manufacturing requires designing adaptable and flexible systems and anticipating runtime changes at design time. In this article, we describe the engineering of (self-)adaptiveness of manufacturing systems by presenting definitions, significant challenges, and promising solution approaches. We limit the scope of work on designing and realizing such manufacturing systems based on reference architectures, self-organization, and knowledge-based reconfiguration.

Prognostic and health management for adaptive manufacturing systems with online sensors and flexible structures

Real-time monitoring and accurate predictions of machine failures are important in maintenance decision-making. Traditional policies using population-specific reliability characteristics cannot represent degradation processes of individual machines, thus result in less accurate predictions of time-to-failure (TTF). Besides, most of the existing maintenance policies focus on a manufacturing system with its fixed system structure, which means the system is designed with limited flexibility. Nowadays, the flexible structure of an adaptive manufacturing system can be adjustable to meet various product types and changeable market demands. In this paper, we try to fill these gaps and develop a prognostic and health management (PHM) framework for manufacturing systems with online sensors and flexible structures. We integrate a Bayesian updating prognostic model using sensor-based degradation information for computing each machine’s TTFs, with an opportunistic maintenance policy handling flexible system structures for optimizing the maintenance scheduling. This enables the dynamic prognosis updating, the notable cost reduction, and the rapid decision making for adaptive manufacturing systems.

Master assembly network for alternative assembly sequences

The fourth industrial revolution (I 4.0) is paving the way for change in manufacturing systems. A logical enabler for dynamic and adaptive manufacturing systems including smart automated guided vehicles (AGVs) is presented. It can respond to requests for changing operations sequences received digitally or via distributed sensors, and change the processing route according to pre-planned flow sequences and pre-determined alternatives. A novel method for generating a master assembly network with alternative sequences based on legacy assembly data for a product family is developed. A master assembly network is a generic multiple alternative assembly sequences for a group of product variants belonging to a family where they share some parts and have common product structure. The assembly network with alternative sequences for a new variant is extracted from the master assembly network. These alternative sequences increase the flexibility and adaptability of the assembly system to handle workshop disruptions such as change orders, machine breakdowns and tool failures. The developed method is inspired by the phylogenetic networks used in biology, namely the soft-wired galled network. A Genetic Algorithm is developed to generate the master assembly network that summarizes a set of conflicting rooted assembly sequence trees. A family of three control valves is used as a case study. The proposed method can be utilized in any manufacturing system that use alternative assembly sequence including those utilizing smart AGVs in and Industry 4.0 dynamic environment. The developed method decreases the time and cost of introducing a new product variant as well as increases the responsiveness of the manufacturing system.

Automated work cycle classification and performance measurement for manual work stations

The increasing demand for highly customized products requires flexible, reactive and adaptive manufacturing systems. Accurate and up-to-date information about the processes is a strict requirement to meet these needs. Real-time data capturing technologies, such as RFID, have already been used for some years in manufacturing environments, mainly for inventory management, planning and quality control. However, these systems fail to generate information on the performance of the operator in the system. This paper presents a video-based system that automates the analysis of manual assembly line work stations and generates near real-time information to support workers in their pursuit of continuous improvement. A work cycle classification method was developed to detect anomalous and problematic situations in the work flow. Besides the classification of work cycles, the method also generates performance indicators to analyze the performance of the operator in the system. These performance indicators are visualized in an operational dashboard, which reveals the improvement potential of the work station.

Implementing Information Technologies and Operational Excellence: Planning, emergence and randomness in the survival of adaptive manufacturing systems

We explore the why and how of synergies and dysfunctions between Information Technologies (IT) and Operational Excellence (OE). Moving away from the type of analyses that focus on the starting-point and the final stage, we address the evolution of the complex factors involved in the joint deployment of IT and OE that try to transform manufacturing systems. Evidence stems from a longitudinal case study over a period of 11 years (2003–2014) in one of the world’s largest canning factories. Combining quasi-experiments at the systems level with time series data, we compare the evolution of IT implementation in a Lean environment with 17 different assembly lines and 1,100 workers. Thus, building on an evolutionary framework, the paper highlights the simultaneity of planned, random and emergent components in mutations; the influence of the rhythm and sequence of reforms in the construction of capabilities; and the effect of environmental evolution on the type of mutations passed on over time.

A framework to design a human-centred adaptive manufacturing system for aging workers

The so-called smart manufacturing systems (SMS) combine smart manufacturing technologies, cyber-physical infrastructures, and data control to realize predictive and adaptive behaviours. In this context, industrial research focused mainly on improving the manufacturing system performance, almost neglecting human factors (HF) and their relation to the production systems. However, in order to create an effective smart factory context, human performance should be included to drive smart system adaptation in efficient and effective way, also by exploiting the linkages between tangible and intangible entities offered by Industry 4.0. Furthermore, modern companies are facing another interesting trend: aging workers. The age of workers is generally growing up and, consequently, the percentage of working 45–64years old population with different needs, capabilities, and reactions, is increasing. This research focuses on the design of human-centred adaptive manufacturing systems (AMS) for the modern companies, where aging workers are more and more common. In particular, it defines a methodology to design AMS able to adapt to the aging workers’ needs considering their reduced workability, due to both physical and cognitive functional decrease, with the final aim to improve the human-machine interaction and the workers’ wellbeing. The paper finally presents an industrial case study focusing on the woodworking sector, where an existing machine has been re-designed to define a new human-centred AMS. The new machine has been engineered and prototyped by adopting cyber-physical systems (CPS) and pervasive technologies to smartly adapt the machine behaviour to the working conditions and the specific workers’ skills, tasks, and cognitive-physical abilities, with the final aim to support aging workers. The achieved benefits were expressed in terms of system usability, focusing on human-interaction quality.

Optimized Adaptive Scheduling of a Manufacturing Process System with Multi-skill Workforce and Multiple Machine Types: An Ontology-based, Multi-agent Reinforcement Learning Approach

The impetus for an interconnected, efficient, and adaptive manufacturing system, as advocated by the Industry 4.0 revolution, together with the latest developments in information technology, advanced manufacturing has become a prominent research topic in recent years. One critical aspect of advanced manufacturing is how to incorporate real-time demand information with a manufacturer's resource information, including workforce data and machine capacity and condition information, among others, to optimally schedule manufacturing processes with multiple objectives. In general, optimized manufacturing scheduling is a non-deterministic polynomial-time hard problem. Due to the complexity, scheduling presents a number of challenges to find the best possible solutions. This research proposes an ontology-based framework to formally represent a synchronized, station-based flow shop with a multi-skill workforce and multiple types of machines. Based on the ontology, this research develops a multi-agent reinforcement learning approach for the optimal scheduling of a manufacturing system of multi-stage processes for multiple types of products with various machines and a multi-skilled workforce. By employing a learning algorithm, this approach enables real-time cooperation between the workforce and the machines, and adaptively updates production schedules according to dynamically changing real-time events.

Towards a griddable distributed manufacturing system with augmented reality interfaces

Rapidly changing demand and mass customization require highly flexible and adaptive manufacturing systems. Manufacturing operations have evolved in order to keep up by organizing themselves into smaller units of specialized production processes that are combined in different ways to create different products. Human workers are integral in the manufacturing systems and they too must be flexible and adaptive. This paper describes an augmented reality manufacturing system that aims to greatly improve the information perception of the different types of workers in a manufacturing facility and to make interaction with manufacturing software natural and efficient. In this approach, traditionally paper-based and computer-based tasks are augmented to the workers' interactions in the environment. The system is distributed and modular as the different functions of CAD/CAM software are provided by individual physical or virtual objects in the environment or by a combination of them working cooperatively. This modularity allows the individual resources and facilities to be linked via internet onto a manufacturing grid with universally-accessible augmented reality interfaces to their services. Framework for ubiquitous AR environments.Application of the framework to a local manufacturing environment.Application of the framework to manufacturing grid.

Adaptive Manufacturing System

A system and method for processing a flexible part (10) comprising holding the flexible part (10) securely in an unconstrained position using a holder (102); and controlling a positioner (152, 153) to process the flexible part (10) based on a comparison of a shape and/or position of the flexible part (10) in the unconstrained position with design specifications of the part (10') not in the unconstrained position.

An Automated Approach for Adaptive Control Systems

The paper deals with adaptive manufacturing systems to be composed of various machines for optimal productions of jobs. The authors assume three types of constraints to be respected: (1) deals with the system’s performance which is related to productivity rates and should be kept acceptable as much as possible, (2) deals with the energy consumption which should be kept stable or minimal, and (3) deals with emissions of CO2 which should be kept minimal for future systems. They define a reconfiguration scenario as any operation allowing addition and/or removals of jobs and/or machines to/from production lines for safety when faults occur. To meet all fixed constraints, the authors propose an agent-based architecture where an intelligent software agent is defined to evaluate the whole system’s architecture after any reconfiguration scenario, and to define useful technical solutions when constraints are violated. It suggests run-time modifications of productivity rates and/or realization times of jobs and/or also removals of some jobs according to well-defined constraints. The users supervising the production lines can dynamically accept or reject these solutions according to production business strategies and also some other constraints. The authors apply the paper’s contribution to a formal drilling manufacturing platform, and to the footwear factory of ITIA-CNR.