On the configuration of supply chains for assemble-to-order products: Case studies from the automotive and the CNC machine building sectors

The product customisation trend has an unprecedented impact on manufacturing companies, as the ever-increasing number of product variants and the enlarged pool of cooperating partners vastly increase the feasible alternative supply chain configurations. In terms of decision theory, this is translated to enormous search spaces. For tackling these NP-hard problems, metaheuristic optimisation methods are utilised, which provide a trade-off between the quality of solutions and the computation time. This research work describes the modelling and solving of two supply chain configuration problems using the Simulated Annealing and Tabu Search methods. The performance of the identified solutions in terms of optimisation of multiple conflicting criteria, is compared against the results obtained from a custom Intelligent Search Algorithm and an Exhaustive enumerative method. The algorithms are developed into a web-based software platform. The approach is validated through real life applications to case studies from the automotive and CNC laser welding machine building industries.     

Automated driving for individualized sheet metal part production—A neural network approach

The manufacturing of individualized sheet metal components is one of the most important issues in industrial sheet metal working. Incremental forming methods, in particular driving, offer the opportunity for achieving this objective. However, these manual processes are very difficult to automate, as a result of their complexity and user interactivity. To resolve this problem, a knowledge-based approach is presented, which utilizes a special type of driving process. Initially, a neural network architecture is established which delivers manufacturing strategies allowing part production for simple component shapes. After providing a method for training data generation, training sessions are carried out. Strategies, computed by trained networks, are adopted for processing sheet blanks which are used for evaluating the framework. Finally, the developed procedure is generalized, and a concept is designed which allows a transfer, in order to facilitate the production of arbitrary individualized sheet metal parts.     

An integrated modelling framework to support manufacturing system diagnosis for continuous improvement

This paper proposes an integrated modelling framework for the analysis of manufacturing systems that can increase the capacity of modelling tools for rapidly creating a structured database with multiple detail levels and thus obtain key performance indicators (KPIs) that highlight possible areas for improvement. The method combines five important concepts: hierarchical structure, quantitative/qualitative analysis, data modelling, manufacturing database and performance indicators. It enables methods to build a full information model of the manufacturing system, from the shopfloor functional structure to the basic production activities (operations, transport, inspection, etc.). The proposed method is based on a modified IDEF model that stores all kind of quantitative and qualitative information. A computer-based support tool has been developed to connect with the IDEF model, creating automatically a relational database through a set of algorithms. This manufacturing datawarehouse is oriented towards obtaining a rapid global vision of the system through multiple indicators. The developed tool has been provided with different scorecard panels to make use of KPIs to decide the best actions for continuous improvement. To demonstrate and validate both the proposed method and the developed tools, a case study has been carried out for a complex manufacturing system.     

Agile MPC system linking manufacturing and market strategies

Increasing complexity and interdependency in manufacturing enterprises require an agile manufacturing paradigm. This paper considers a dynamic control approach for linking manufacturing strategy with market strategy through a reconfigurable manufacturing planning and control (MPC) system to support agility in this context. A comprehensive MPC model capable of adopting different MPC strategies through distributed controllers of inventory, capacity, and WIP is presented. A hierarchical supervisory controller (referred to as decision logic unit, DLU) that intakes the high-level strategic market decisions and constraints together with feedback of the current manufacturing system state (WIP, production, and inventory levels) and optimally manages the distributed controllers is introduced. The DLU architecture with its three layers and their different functionalities is discussed showing how they link the higher management level to the operational level to satisfy the required demand. A case study for an automatic PCB assembly factory is implemented to demonstrate the applicability of the whole approach. In addition, a comparative cost analysis study is carried out to compare between the developed agile MPC system and classical-inventory- and capacity-based MPC policies in response to different demand patterns. Results showed that the developed agile MPC policy is as cost effective as the inventory-based MPC policy in demand patterns with steady trends, as cost effective as capacity-based MPC in turbulent demand patterns, and far superior than both classical MPC polices in mixed-demand patterns.     

一种新的启发式边排序策略及其性能分析

网络可靠度BDD分析方法的计算复杂度与BDD尺度线性相关,而BDD尺度严重依赖边排序质量。由于求解最优边排序是一个NP问题,在实际应用中,通常采用启发式边排序策略如BFS（Breadth First Search）和DFS（Depth First Search）。针对边排序问题,从分析基于边界集（Boundary Set）的BDD构建方法BDD BS出发,将边界集思想应用于边排序过程,提出了一种新的启发式边排序策略。性能分析和大量实验表明,新设计的边排序策略性能优于经典的DFS和BFS策略,该结果为网络可靠度BDD分析方法在大规模网络中的应用拓展了新的空间。     

Reinforcement learning for resource allocation in LEO satellite networks.

In this paper, we develop and assess online decision-making algorithms for call admission and routing for low Earth orbit (LEO) satellite networks. It has been shown in a recent paper that, in a LEO satellite system, a semi-Markov decision process formulation of the call admission and routing problem can achieve better performance in terms of an average revenue function than existing routing methods. However, the conventional dynamic programming (DP) numerical solution becomes prohibited as the problem size increases. In this paper, two solution methods based on reinforcement learning (RL) are proposed in order to circumvent the computational burden of DP. The first method is based on an actor-critic method with temporal-difference (TD) learning. The second method is based on a critic-only method, called optimistic TD learning. The algorithms enhance performance in terms of requirements in storage, computational complexity and computational time, and in terms of an overall long-term average revenue function that penalizes blocked calls. Numerical studies are carried out, and the results obtained show that the RL framework can achieve up to 56% higher average revenue over existing routing methods used in LEO satellite networks with reasonable storage and computational requirements.     

A hybrid method based on genetic algorithm and dynamic programming for solving a bi-objective cell formation problem considering alternative process routings and machine duplication

Cell formation is one of the first and most important steps in designing a cellular manufacturing system. It consist of grouping parts with similar design features or processing requirements into part families and associated machines into machine cells. In this study, a bi-objective cell formation problem considering alternative process routings and machine duplication is presented. Manufacturing factors such as part demands, processing times and machine capacities are incorporated in the problem. The objectives of the problem include the minimization of the total dissimilarity between the parts and the minimization of the total investment needed for the acquisition of machines. A normalized weighted sum method is applied to unify the objective functions. Due to the computational complexity of the problem, a hybrid method combining genetic algorithm and dynamic programming is developed to solve it. In the proposed method, the dynamic programming is implemented to evaluate the fitness value of chromosomes in the genetic algorithm. Computational experiments are conducted to examine the performance of the hybrid method. The computations showed promising results in terms of both solution quality and computation time.     

Topology optimization for hybrid additive-subtractive manufacturing

Hybrid additive-subtractive manufacturing is gaining popularity by making full use of geometry complexity produced by additive manufacturing and dimensional accuracy derived from subtractive machining. Part design for this hybrid manufacturing approach has been done by trial-and-error, and no dedicated design methodology exists for this manufacturing approach. To address this issue, this work presents a topology optimization method for hybrid additive and subtractive manufacturing. To be specific, the boundary segments of the input design domain are categorized into two types: (i) Freeform boundary segments freely evolve through the casting SIMP method, and (ii) shape preserved boundary segments suppress the freeform evolvement and are composed of machining features through a feature fitting algorithm. Given the manufacturing strategy, the topology design is produced through additive manufacturing and the shape preserved boundary segments will be processed by post-machining. This novel topology optimization algorithm is developed under a unified SIMP and level set framework. The effectiveness of the algorithm is proved through a few numerical case studies.     

Statics and dynamics of cognitive and qualitative matchmaking in task fulfillment

Two main ingredients related to successful task performance are cognition and quality. Supply and demand of these concepts for knowledge intensive tasks are studied in this paper to fuel successful task fulfillment. Cognitive characteristics are supplied by actors performing tasks. Organizational developments such as growing complexity and increasing customer orientation may increase cognitive load. Stakeholders of tasks have quality requirements. These requirements may be affected if actors experience an increase in cognitive load. It is observed that knowledge intensive tasks demand cognitive characteristics and supply quality factors. Actors supply cognition and stakeholders demand quality. The gap between supply and demand can be bridged by introducing several models. These models consist of a matchmaking framework, conceptual models, and dynamic models. The matchmaking framework shows how supply and demand of cognitive characteristics or quality factors can be matched. Relations and roles of the concepts involved in task fulfillment are mapped out by the conceptual models. The dynamic models show causes that have effects on the supply of cognitive characteristics and the level of quality. These insights in the relations and dependencies between cognition and quality increase our understanding of the key concepts for successful task fulfillment.     

Neuro-genetic approach to optimize parameter design of dynamic multiresponse experiments

Engineers have widely applied the Taguchi method, a traditional approach for robust experimental design, to a variety of quality engineering problems for enhancing system robustness. However, the Taguchi method is unable to deal with dynamic multiresponse owing to increasing complexity of the product or design process. Although several alternative approaches have been presented to resolve this problem, they cannot effectively treat situations in which the control factors have continuous values. This study incorporates desirability functions into a hybrid neural network/genetic algorithm approach to optimize the parameter design of dynamic multiresponse with continuous values of parameters. The objective is to find the optimal combination of control factors to simultaneously maximize robustness of each response. The proposed approach is based on three stages which (1) use neural networks for constructing a response function model of a dynamic multiresponse system, (2) use exponential desirability functions for evaluating overall performance of a specific factor combination, and (3) use a genetic algorithm to optimize parameter design. Effectiveness of the proposed approach is illustrated with a simulated example. Analysis results reveal that the approach has higher performance than the traditional experimental design.     

Framework for Assessing Robotic Dexterity within Flexible Manufacturing

With growing demand for flexibility in manufacturing processes, interest in dexterous industrial robots is increasing. To facilitate benchmarking, and to assess the suitability of these robots for flexible manufacturing tasks, there is a need to develop new methods of capturing the relevant performance characteristics of industrial robots. This research aims to show that the Boothroyd-Dewhurst (B-D) Design-For-Assembly method, an established method for optimizing manufacturing processes, can be effectively adopted to form the basis of a comprehensive robotic dexterity assessment within flexible manufacturing. A comparative study is conducted which shows that the B-D classification tables offer the most comprehensive solution due to the range of operations and artifacts considered. Building on these tables, a framework is developed for determining the suitability of a robot system within flexible manufacturing operations. In a sample test-case scenario involving a pick-and-place operation, the framework is shown to produce an accurate estimate of robot performance that can be easily compared to human data. The framework establishes a link between manufacturing operations and robot performance metrics, which addresses the current difficulty in robot integration and highlights the framework’s potential for adoption within flexible manufacturing.     

A framework for intelligent design of manufacturing cells

One of the major thrusts of agile/lean/responsive manufacturing strategies of the twentyfirst century is to introduce advanced information technology into manufacturing. This paper presents a framework for robust manufacturing system design with the integration of simulation, neural networks and knowledge-based expert system tools. An operation/ cost-driven cell design methodology was applied to concurrently consider cell physical design and the complexity of cell control functions. Simulation was exercised to estimate performance measures based on input parameters and given cell configurations. A rulebased expert system was employed to store the acquired expert knowledge regarding the relation between cell control complexities, cost of cell controls, performance measures and cell configuration. Neural networks were applied to predict the cell design configuration and corresponding complexities of cell control functions. Training of neural networks was performed with both forward and backward methods by using the same pair of data sets. Hence, trained neural networks will be able to predict either input or output parameters. This innovative new design methodology was illustrated via a successful implementation exercise resulting in actually acquiring an automated cell at industrial settings. The experience learned from this exercise indicates that the proposed design methodology works well as an effective decision support system for cell designers and the management in determining appropriate cell configuration and cell control functions at the design stage.     

Dynamic power management in energy-aware computer networks and data intensive computing systems

Energy awareness is an important aspect of modern network and computing system design and management, especially in the case of internet-scale networks and data intensive large scale distributed computing systems. The main challenge is to design and develop novel technologies, architectures and methods that allow us to reduce energy consumption in such infrastructures, which is also the main reason for reducing the total cost of running a network. Energy-aware network components as well as new control and optimization strategies may save the energy utilized by the whole system through adaptation of network capacity and resources to the actual traffic load and demands, while ensuring end-to-end quality of service. In this paper, we have designed and developed a two-level control framework for reducing power consumption in computer networks. The implementation of this framework provides the local control mechanisms that are implemented at the network device level and network-wide control strategies implemented at the central control level. We also developed network-wide optimization algorithms for calculating the power setting of energy consuming network components and energy-aware routing for the recommended network configuration. The utility and efficiency of our framework have been verified by simulation and by laboratory tests. The test cases were carried out on a number of synthetic as well as on real network topologies, giving encouraging results. Thus, we come up with well justified recommendations for energy-aware computer network design, to conclude the paper.     

Computer-assisted manufacturing process optimization with neural networks

Today's manufacturing methods are caught between the growing need for quality, high process safety, minimal manufacturing costs, and short manufacturing times. In order to meet these demands, process setting parameters have to be chosen in the best possible way, according to demand on quality. For such optimization it is necessary to represent the processes in a model. Due to the enormous complexity of many processes and the high number of influencing parameters, however, conventional approaches to modelling and optimization are no longer sufficient. In this article it is shown how, by means of applying neural networks for process modelling, even these highly complex interdependencies can be learned. That way both process and quality parameters can be assessed before or during processing. By connecting them with corresponding cost models, it is possible to optimize processes with the help of evolutionary algorithms. Using examples of different manufacturing processes, the possi bilities for process modelling and optimization with neural networks and evolutionary algorithms are demonstrated.     

Concurrent manufacturing of parts and tools for the sheet-metal industry

The manufacturing of complex bent parts can be supported effectively by computer-aided planning methods. Software systems are already available for unfolding, laser cutting and bending sequence determination. The paper focuses on methods that support the design of non-standard bending tools and the flexible manufacturing of such tools using laminated object modelling (LOM) technology. The developed system allows for concurrent planning and manufacturing of bending parts and tools. Within the framework of this system, neural networks are applied for automated tool design.     

Dynamically Integrated Manufacturing Systems (DIMS)—A Multiagent Approach

Manufacturing businesses in today's market are facing immense pressures to react rapidly to dynamic variations in demand distributions across products and changing product mixes. To cope with the pressures requires dynamically integrated manufacturing systems (DIMS) that can manage optimal fulfillment of customer orders while simultaneously considering alternative system structures to suit changing conditions. This paper presents a multiagent approach to DIMS, where production planning and control decisions are integrated with systems reconfiguration and restructure. A multiagent framework, referred to as a hierarchical autonomous agent network, is proposed to model complex manufacturing systems, their structures, and constraints. It allows the hierarchical structures of complex systems to be modeled while avoiding centralized control in classical hierarchical/hybrid frameworks. Subsystems interact heterarchically with product orders to carry out optimal planning and scheduling. An agent coordination algorithm, operating iteratively under the control of a genetic algorithm, is developed to enable optimal planning and control decisions for order fulfillment to be made through interactions between agents. This algorithm also allows the structural constraints of systems to be relaxed gradually during agent interaction, so that planning and control are first carried out under existing constraints, but when satisfactory solutions cannot be found, subsystems are allowed to regroup to form new configurations. Frequently used configurations are detected and evaluated for system restructure. The approach also enables Petri-net models of new system structures to be generated dynamically and the structures to be evaluated through agent-based discrete event simulation.     

A similarity hybrid harmony search algorithm for the Team Orienteering Problem

The Harmony Search (HS) is a metaheuristic algorithm, which is inspired from the composition of music harmonies. The functionality and flexibility of HS contribute to the development of successful methodologies for different kind of scientific problems. The aim of this paper is to propose a variant of the classic HS algorithm in order to provide competitive solutions for the Team Orienteering Problem (TOP). We introduce the Similarity Hybrid Harmony Search (SHHS) algorithm as an alternative and innovative optimization method. The SHHS follows the standard procedure of HS with some modifications and includes a new strategy called “similarity process”. Two versions of the proposed method have been developed, the static version with predefined values for the parameters of the method and the dynamic one with dynamic adjustment of the parameters. The SHHS algorithm is applied to the known benchmark instances of TOP. The dynamic of the algorithm is tested through a complete solution analysis which gives the superiority of the dynamic version compared to the static one. The results of both versions of the proposed algorithm indicate the positive performance against other effective and robust optimization algorithms from the literature.     

A generic stochastic model for supply-and-return network design

This paper presents a generic stochastic model for the design of networks comprising both supply and return channels, organized in a closed loop system. Such situations are typical for manufacturing/re-manufacturing type of systems in reverse logistics. The model accounts for a number of alternative scenarios, which may be constructed based on critical levels of design parameters such as demand or returns. We describe a decomposition approach to this model, based on the branch-and-cut procedure known as the integer L-shaped method. Computational results in an illustrative numerical setting show a consistent performance efficiency of the method. Moreover, the stochastic solution features a significant improvement in terms of average performance over the individual scenario solutions. A modeling and solution methodology as presented here can contribute to the efficient solution of network design models under uncertainty for reverse logistics.     

A novel grouping harmony search algorithm for the multiple-type access node location problem

In this paper we present a novel grouping harmony search algorithm for the Access Node Location Problem (ANLP) with different types of concentrators. The ANLP is a NP-hard problem where a set of distributed terminals, with distinct rate demands, must be assigned to a variable number of concentrators subject to capacity constraints. We consider the possibility of choosing between different concentrator models is given in order to provide service demand at different cost. The ANLP is relevant in communication networks design, and has been considered before within the design of MPLS networks, for example. The approach we propose to tackle the ANLP problem consists of a hybrid Grouping Harmony Search (GHS) algorithm with a local search method and a technique for repairing unfeasible solutions. Moreover, the presented scheme also includes the adaptation of the GHS to a differential scheme, where each proposed harmony is obtained from the same harmony in the previous iteration. This differential scheme is perfectly adapted to the specifications of the ANLP problem, as it utilizes the grouping concept based on the proximity between nodes, instead of being only based on the grouping concept. This allows for a higher efficiency on the searching process of the algorithm. Extensive Monte Carlo simulations in synthetic instances show that this proposal provides faster convergence rate, less computational complexity and better statistical performance than alternative algorithms for the ANLP, such as grouping genetic algorithms, specially when the size of the scenario increases. We also include practical results for the application of GHS to a real wireless network deployment problem in Bizkaia, northern Spain.     

A metamodel-based Monte Carlo simulation approach for responsive production planning of manufacturing systems

Production planning is concerned with finding a release plan of jobs into a manufacturing system so that its actual outputs over time match the customer demand with the least cost. For a given release plan, the system outputs, work in process inventory (WIP) levels and job completions, are non-stationary bivariate time series that interact with time series representing customer demand, resulting in the fulfillment/non-fulfillment of demand and the holding cost of both WIP and finished-goods inventory. The relationship between a release plan and its resulting performance metrics (typically, mean/variance of the total cost and the fill rate) has proven difficult to quantify. This work develops a metamodel-based Monte Carlo simulation (MCS) method to accurately capture the dynamic, stochastic behavior of a manufacturing system, and to allow real-time evaluation of a release plan's performance metrics. This evaluation capability is then embedded in a multi-objective optimization framework to search for near-optimal release plans. The proposed method has been applied to a scaled-down semiconductor fabrication system to demonstrate the quality of the metamodel-based MCS evaluation and the results of plan optimization.