Uncertain Manufacturing Environment Research Articles

Evaluation of Collaborative Robot Sustainable Integration in Manufacturing Assembly by Using Process Time Savings.

Collaborative robots are enablers of flexibility in the current dynamic and uncertain manufacturing environment. Decision making on its implementation requires technical feasibility, involving productivity and workforce implications that should be faced in an integrated perspective in processes where many components of different materials are assembled in products of increasing diversity and complexity. This study introduces two new parametric models for collaborative robotics, formulated in order to evaluate the differential cost of assembly (economic dimension) and the differential income from taxes that supports short-term workforce displacement (social dimension) in cobot implementation. Updated techno-economical parameters are selected for assessing feasibility ranges of application in different production scenarios. Next, the influence curves of productivity gain for a feasible implementation of cobot establish thresholds for decision making under both criteria. The results show the need for productivity gains that are significantly lower in high-wage scenarios than in low-wage scenarios; however, in a joint approach, breakeven productivity gain is always higher for the social dimension threshold than for the economic requirement of cost-effective manufacturing, with a higher gap in low-wage cases. The detailed analysis of a real case study of cobot implementation for assembly demonstrates the practical application of models and potential for future research.

Industrial big-data-driven and CPS-based adaptive production scheduling for smart manufacturing

Smart manufacturing that involves tight integration of the physical system and cyber system is a hot topic in both industry and academia in the era of the Internet and big data. However, the dynamic and uncertain manufacturing environment introduces a significant adaptive issue of production scheduling, which is one of the pivotal tasks for smart manufacturing. This paper focuses on this problem and proposes a closed-loop adaptive scheduling solution based on the Cyber-Physical Production System (CPPS) with four phases: production data acquisition (PDA), dynamic disturbance identification (DDI), scheduling strategy adjustment (SSA), and schedule scheme generation (SSG). In the DDI phase, in view of the disturbance classification, a disturbance identification procedure based on CPPS monitoring is studied to ensure real-time response. In the SSA phase, an industrial big-data-driven scheduling strategy adjustment method is proposed, which consists of GA-based offline knowledge learning and KNN-based online adjustment, to enhance the system adaptability. We apply and verify the proposed adaptive scheduling solution on an experimental semiconductor manufacturing system, and the results demonstrate that the proposed method outperforms the dynamic scheduling method in terms of multiple objectives under different disturbance levels.

Multi-perspective collaborative scheduling using extended genetic algorithm with interval-valued intuitionistic fuzzy entropy weight method

Scheduling multiple heterogeneous tasks in a manufacturing system to satisfy customized requirements becomes challenging, especially in uncertain manufacturing environment. In cloud manufacturing, a serious problem is to schedule multiple heterogeneous tasks to balance the benefit conflicts among customers, manufacturing enterprises, and manufacturing platform comprehensively. Therefore, this study formulates the multi-task scheduling problem mathematically as a new fuzzy mixed-integer linear programming (FMILP) model based on multi-perspective collaborative optimization and fuzzy set theory. To solve the FMILP model, an extended genetic algorithm (EGA) with the interval-valued intuitionistic fuzzy entropy weight (IVIFEW) method is proposed. The IVIFEW method is adapted to obtain the preference of QoS attributes and task priority. In addition, the basic genetic algorithm is improved by integrating a migration operator, local search, and restart strategy to maintain the diversity of population and enhance the exploitation ability. A suitable parameter combination of EGA is found in a series of experiments based on the Taguchi method. The experimental results demonstrate that the proposed EGA solves the FMILP model effectively, providing better optimal solutions compared with the baseline algorithms.

A Hybrid Cooperative Coevolution Algorithm for Fuzzy Flexible Job Shop Scheduling

Flexible scheduling is one of the most significant core techniques for intelligent manufacturing systems. Realization of an optimized schedule through flexible resources assignment is critical to the application and popularization of flexible scheduling, especially in uncertain manufacturing environments. In this paper, we consider flexible job shop scheduling with uncertain processing time represented by fuzzy numbers, which is named fuzzy flexible job shop scheduling. We propose an effective hybrid cooperative coevolution algorithm (hCEA) for the minimization of fuzzy makespan. The hCEA combines particle swarm optimization with the genetic algorithm to improve the convergence ability. A parameter self-adaptive strategy is applied to the problems with different scale effectively as well. Five benchmarks and three large-scale problems with fuzzy processing time are adopted to test the hCEA. Computational results show that the hCEA performs better than the existing methods from the literature.

An internet-of-things-based production logistics optimisation method for discrete manufacturing

ABSTRACTTimely supply of production material is an important prerequisite for stable operation of discrete manufacturing systems. The difficulties of production logistics (PL) planning are increased due to the uncertainty and dynamicity in production environment. The Internet of things (IoT) provides a reliable solution for monitoring dynamic manufacturing process and obtaining real-time information. Under the uncertain manufacturing environment, this paper focusses on a PL optimisation method driven by real-time data. First, considering the uncertainty of material demands time caused by production fluctuation, the mathematical scheduling model with fuzzy time windows is established to minimise the distribution cost. Later, on the basis of the proposed two-stage operation and optimisation mechanism, an improved ant colony algorithm is designed, which introduces the factors of satisfaction degree and time windows width into state transfer rules and improves the dynamic adjustment strategy of pheromone. Finally, in a machining workshop, a case study is conducted based on the construction of real-time sensing and positioning manufacturing environment. Several numerical experiments are performed to verify the feasibility of the proposed method.

Optimisation of net profit with uncertain inputs in manufacturing environments by integration of neural networks, genetic algorithm and fuzzy regression

This paper presents an integrated artificial fuzzy regression, neural network (ANN) and genetic algorithm (GA) for optimisation of profit with uncertain inputs. Generally, truncations of α has been used to study the fuzzy regression model. In this paper, fuzzy regression is accomplished by the fuzzy neural networks and the necessary neural nets training is proposed by the fuzzy numbers which is based on genetic algorithm. The proposed neural net learning method based on GA is claimed to be a better substitute because of its higher efficiency. To show the applicability and superiority of the proposed approach an actual case study (manufacturer of aluminium heater) is presented, applied and discussed for the improved fuzzy regression by the integrated neural network and genetic algorithm. This is the first study that integrates fuzzy regression, GA and ANN for optimisation of net profit in an uncertain manufacturing environment.

FUZZY LOGIC BASED UNCERTAINTY REPRESENTATION AND SIMULATION IN A FLEXIBLE ASSEMBLY SYSTEM

This paper provides intelligent simulation approach using manufacturing uncertainty in the form of simulation intelligence to improve the performances of manufacturing production system. It shows how simulation can be used to evaluate alternative designs in an uncertain manufacturing environment. Fuzzy rule-based manufacturing uncertainties are addressed in this study. A combination of product mix and production volume is analyzed using intelligent simulation model for an optimal production design. The intelligent simulation approach would improve the modeling accuracy in terms of more realistic presentation of uncertain activities. The proposed intelligent simulation modeling shows significant improvement and validated with real life application.

EFFICIENT MATERIAL ALLOCATIONS IN HIGH-MIX LOW-VOLUME MANUFACTURING

Today's globalization market drives industries toward increased expectations on lean production and quick response to the customers. These expectations have put industries under pressure to become more agile under highly dynamic market and uncertain manufacturing environment. An efficient material allocation becomes critical due to dynamics and uncertainty in the manufacturing systems. This research addresses the optimization needs for effective material allocations in high-mix low-volume manufacturing. Multi-constraint based genetic algorithm and intelligent simulation are used for efficient material allocation. The proposed effective material allocation is validated through real-world application in electronics manufacturing.

Formal design and analysis of a hybrid supervisory control structure for Virtual Production Systems

Virtual Production Systems (VPSs) is a dynamic paradigm for production resources structure, which was proposed to cope with changing and uncertain manufacturing environment. However, the high performance of VPSs relies on a high-quality control in practice. Motivated by both local quick responses and global optimization, a supervisory control structure based on autonomous and coordination mechanisms is proposed for VPSs in this paper. This hybrid supervisory control structure is formally designed and analysed with timed automata on the platform of an integrated tool named UPPAAL. First, the Discrete Events Dynamic Systems (DEDS) model of VPSs is established. Then autonomous and coordinated supervisory controllers are further designed under the guidance of heuristic scheduling rules, such that the desired characteristics of both performance (production flow and time) and activity (overflow-, conflict- and deadlock-free) can be ensured. Finally, system analysis, including deadlock-free analysis, calculation and simulation of time-optimal scheduling, is made through verification. A case study is made to illustrate that control and scheduling can be well integrated into this hybrid supervisory control structure in practice.

An integrated system for managing ship repair operations

The global competition pressure drives manufacturing companies to be lean, virtual and global. However many manufacturing companies, such as ship repair yards, are handicapped in marketing competition owing to their complex and uncertain manufacturing environment, resulting in problems related to scheduling, control and capacity planning of their manufacturing operations. The approach discussed in this paper is designed to support the management of a ship repair yard by integrating in an open and flexible system a number of critical business functions with production planning, scheduling and control. This approach is implemented in a software system fully developed in Java and designed by using UML. The proposed software system operates at a ship repair yard in a pilot mode, concurrently with the existing working practice. With the use of the new concept for production planning activities, the managers of the shipyard are assisted to produce overall schedules in a more efficient way in order to allocate work in the various departments.

Dynamic scheduling of stochastic jobs on a single machine

In this paper, the problem of minimizing the expected number of tardy jobs on a single machine is investigated. Jobs having stochastic processing times and deterministic due dates arrive randomly. The objectives of this research are to evaluate the role of the variance of processing time and to develop a dynamic scheduling policy, which is simple and robust, so that it can be used in rapidly changing and uncertain manufacturing environments. A heuristic based on a myopically optimal solution is developed. The sufficient conditions for the optimality and the worst case analysis are also provided. Computational study validates its effectiveness by comparison with optimal solutions.

Flexible reasoning using sensible agent-based systems: A case study in job flow scheduling

Planning and control systems for highly dynamic and uncertain manufacturing environments require adaptive flexibility and decision-making capabilities. Modern distributed manufacturing systems assess the utility of planning and executing solutions for both system goals (e.g. minimize manufacturing production time for all parts or minimize WIP) and local goals (e.g. expedite part A production schedule or maximize machine X utilization). Sensible Agents have the ability to alter their autonomy levels to choose among a set of decision models in order to handle the differences between local and system goals. In this paper, Sensible Agents are applied to a production planning and control problem in the context of job shop scheduling and decision model theory. Sensible Agents provide for trade-off reasoning mechanisms among system and local utilities that are flexible and responsive to an agent's abilities, situational context and position in the organizational structure of the system.

A framework for classifying flexibility types in manufacturing

In today's dynamic and uncertain manufacturing environment, flexibility is one of the most sought-after properties for manufacturing systems. Despite this interest, flexibility remains poorly understood in theory and utilised in practice. One of the underlying reasons for this is the fact that there is no general agreement on how to define flexibility: over 70 terms can be found in the literature. This paper is concerned with developing a unified framework for classifying the various types of flexibilities found in manufacturing. The framework is an attempt to provide both a mechanism for classifying existing flexibility types and understanding their relationships, and a foundation for future efforts in defining flexibility types and measures. The framework is developed based upon the architectural concepts of a ‘basic mechanism model’ and system design activity, and a system/environment model. Examples of how some existing flexibility types may be classified according to the framework are given.