Multi-objective cell formation and production planning in dynamic virtual cellular manufacturing systems

This article presents a fuzzy goal programming-based approach for solving a multi-objective mathematical model of cell formation problem and production planning in a dynamic virtual cellular manufacturing system. In a dynamic environment, the product mix and part demand change over a planning horizon decomposed into several time periods. Thus, the cell formation done for one period may be no longer efficient for subsequent periods and hence reconfiguration of cells is required. Due to the variation of demand and necessity of reconfiguration of cells, the virtual cellular manufacturing (VCM) concept has been proposed by researchers to utilise the benefits of cellular manufacturing without reconfiguration charges. In a VCM system, machines, parts and workers are temporarily grouped for one period during which machines and workers of a group dedicatedly serve the parts of that group. The only difference of VCM with a real CM is that machines of the same group are not necessarily brought to a physical proximity in VCM. The virtual cells are created periodically depending on changes in demand volumes and mix, as new parts accumulate during a planning horizon. The major advantage of the proposed model is the consideration of demand and part mix variation over a multi-period planning horizon with worker flexibility. The aim is to minimise holding cost, backorder cost and exceptional elements in a cubic space of machine–part–worker incidence matrix. To illustrate the applicability of the proposed model, an example has been solved and computational results are presented.     

Fuzzy neural network based yield prediction model for semiconductor manufacturing system

Accurate die yield prediction is very useful for improving yield, decreasing cost and maintaining good relationships with customers in the semiconductor manufacturing industry. To improve prediction accuracy of die yield, a novel fuzzy neural networks based yield prediction model is proposed in which the impact factors of yield and critical electrical test parameters are considered simultaneously and are taken as independent variables. The mapping between these independent variables and yield is constructed in the fuzzy neural network (FNN). The lineal regression between FNN-based yield predicting output and actual yield demonstrates the effectiveness of the proposed approach by historical experimental data of semiconductor fabrication line in Shanghai. The comparison experiment verifies the proposed yield prediction method improves on three traditional yield prediction methods with respect to prediction accuracy.     

Using a multi-agent system to optimise resource utilisation in multi-site manufacturing facilities

Due to cost economies and better serving the global market, many enterprises expanded their manufacturing environment from a localised, single-site facility to more globalised, multi-site facilities. In order to take advantage of operating multi-site facilities, it is vital to make optimisation decisions of resource utilisation as if these facilities situated across different geographical locations are one integrated facility and take into account of the extended multi-site constraints and variables. This paper proposes a multi-agent system, using its characteristics of autonomy and intelligence, to integrate process planning and production scheduling across different facilities, so as to secure the most efficient and cost-effective plan and schedule to meet the demand. A currency-based agent iterative bidding mechanism is developed to facilitate the co-ordination of agents to achieve the goal. A genetic algorithm is employed to tune the currency values for agent bidding. In this paper, a case study is used for simulation in order to demonstrate the effectiveness and performance of the proposed agent system.     

Optimal production control policy in unreliable batch processing manufacturing systems with transportation delay

This paper considers the problem of production planning of unreliable batch processing manufacturing systems. The finished goods are produced in lots, and are then transported to a storage area in order to continuously meet a constant demand rate. The main objective of this work is to jointly determine the optimal lot sizing and optimal production control policy that minimise the total expected cost of inventory/backlog and transportation, over an infinite time horizon. The decision variables are the lot sizing and the production rate. The problem is formulated with a stochastic dynamic programming model and the impulse control theory is applied to establish the Hamilton–Jacobi–Bellman (HJB) equations. Based on a numerical resolution of the HJB equations, it is shown that the optimal control policy is governed by a base stock policy for production rate control and economic lot size for batch processing. A thorough analysis and practical issues are addressed with a simulation-based approach. Thus, a combined discrete–continuous simulation model is developed to determine the optimal parameters of the proposed policy when the failure and repair times follow general distributions. The results are illustrated with numerical examples and confirmed through sensitivity analysis.     

A matrix-based Bayesian approach for manufacturing resource allocation planning in supply chain management

Nowadays, the supply chain of manufacturing resources is typically a large complex network, whose management requires network-based resource allocation planning. This paper presents a novel matrix-based Bayesian approach for recommending the optimal resource allocation plan that has the largest probability as the optimal selection within the context specified by the user. A proposed matrix-based representation of the resource allocation plan provides supply chain modelling with a good basis to understand problem complexity, support computer reasoning, facilitate resource re-allocation, and add quantitative information. The proposed Bayesian approach produces the optimal, robust manufacturing resource allocation plan by solving a multi-criteria decision-making problem that addresses not only the ontology-based static manufacturing resource capabilities, but also the statistical nature of the manufacturing supply chain, i.e. probabilities of resource execution and resource interaction execution. A genetic algorithm is employed to solve the multi-criteria decision-making problem efficiently. We use a case study from manufacturing domain to demonstrate the applicability of the proposed approach to optimal manufacturing resource allocation planning.     

Application and modeling of resource service trust-QoS evaluation in manufacturing grid system

Resource scheduling is the bottleneck in MGrid; the current research on resource scheduling strategy is mainly based on resource performance-QoS (quality of service), but the factor of trust-QoS is ignored, which will result in unreasonable and unpractical scheduling results. In order to enhance the validity and success rate of resource scheduling in a manufacturing grid (MGrid) system, provide high credible resource service abilities and results to the user, the concept of resource service trust-QoS is presented; trust-QoS was introduced into MGrid resource service scheduling and the important roles it plays emphasized. The trust problems existing in the resource service transaction between resource service demanders (RSD) and resource service providers (RSP) are put forward. The trust-QoS relationship model which is capable of capturing a comprehensive range of trust relationships which exist in the MGrid system is put forward. Then a two-layer resource service trust-QoS evaluation model is put forward, including an intra-domain trust-QoS evaluation model and an inter-domain trust-QoS evaluation model. The quantitative evaluating algorithms of trust-QoS degree value are proposed and described in detail, as well as the value of real-time and dynamic updating algorithms of trust-QoS degree. Finally, an application prototype, namely MBRSPP-MGrid, is developed. The experimental results of the case study show that the proposed models and algorithms are effective and useful.     

Relative impact of different ERP forms on manufacturing organisations: an exploratory analysis of a global manufacturing survey

There are many types of enterprise resource planning (ERP) systems, ranging from very large and very functional vendor products such as those provided by SAP and Oracle, through in-house systems, and smaller vendor products. Thus, there is a substantial range of enterprise computing support available for manufacturing organisations and their manufacturing planning and control. The Global Manufacturing Research Group (GMRG) has collected a systematic survey of manufacturing organisations around the world, providing a picture of manufacturing operations. We have taken GMRG data and organised it around seven levels of ERP functionality, and analysed these data to determine effectiveness in terms of how ERP systems are used in global manufacturing firms, their role in accomplishing manufacturing planning and control, the relationship between ERP forms and data-management practices, the satisfaction firms have across ERP forms, and finally, relative perceived benefits and costs across ERP forms.     

A spread-sheet model for efficient production and scheduling of a manufacturing line/cell

In this paper we develop an efficient spread-sheet production planning/scheduling model for a resource-constraint production line or a manufacturing cell that produces several products but one at a time with significant changeover time and changeover cost. There are also management and physical constraints related to the operating hours, production capacity and amount of inventory allowed. The production line/cell supplies several products to customers who pull the products according to their own operating policy (working hours) that may be different from manufacture's operating hours. We also show several real-world applications and highlight the benefits and merits of the model.     

Design of cellular manufacturing system with quasi-dynamic dual resource using multi-objective GA

A new concept is presented in this paper of quasi-dynamic cell formation for the design of a cellular manufacturing system, based on analysing the fact that static and dynamic cell formation could not reflect the real situation of a modern cellular manufacturing system. Further, workforce resources are integrated into quasi-dynamic cell formation and thus a quasi-dynamic dual-resource cell-formation problem is proposed. For solving this problem, this paper first establishes a non-linear mixed integer programming model, where inter-cell and intra-cell material cost, machine relocation cost, worker operation time, loss in batch quality and worker salary are to be minimised. Then, a multi-objective GA is developed to solve this model. Finally, a real life case study is conducted to validate the proposed model and algorithm. The actual operation results show that the case enterprise significantly decreases its material handling cost and workforce number and obviously increases its product quality after carrying out the obtained scheme.     

焊缝金属奥氏体分解温度预测研究

根据182组实测焊缝金属奥氏体分解温度值,分别采用线性回归方法、非线性回归方法和人工神经网络技术建立了奥氏体分解温度的预测公式或模型。结果表明:线性回归公式难以准确体现各因素与奥氏体分解温度之间的关系,引入Mo指数和ln(t8/3)函数,预测精度有所提高;考虑了各因素之间交互作用的神经网络模型预测精度高于线性和非线性回归公式的,更适合于奥氏体分解温度预测研究。     

A framework for assessing the use of lean production practices in manufacturing cells

This study introduces a framework for assessing the use of lean production (LP) practices in manufacturing cells (MCs). The development of the framework included four stages: (a) defining LP practices applicable to MC, based on criteria such as the inclusion of practices that workers could observe, interact with and use on a daily basis; (b) defining attributes for each practice, emphasising the dimensions which were typical of their implementation in LP environments; (c) defining a set of evidence and sources of evidence for assessing the existence of each attribute–the sources of evidence included direct observations, analysis of documents, interviews and a feedback meeting to validate the assessment results with company representatives; (d) drawing up a model of the relationships among the LP practices, based on a survey with LP experts. This model supports the identification of improvement opportunities in MC performance based on the analysis of their interfaces. A case study of an MC from an automobile parts supplier is presented to illustrate the application of the framework.     

Parts-mix analysis in recycling-oriented manufacturing systems for multiple-item production

One solution to environmental problems is to construct a recycling-oriented manufacturing system, in which waste products after consumption are collected, and reusable parts in the collected products are employed to produce new products. A parts-mix in the manufacturing system for multiple-item production is analysed by taking into consideration the merits and demerits of the usage of common parts over multiple products and special parts utilized for individual products. The parts-mix problem is formulated to minimize the life-cycle cost, which consists of production costs and recycling costs. The property of the optimal parts-mix is clarified, and a decision procedure to solve the parts-mix is developed. The procedure decides whether the common parts are effective or not.     

On the stability and bullwhip effect of a production and inventory control system

This paper focuses on the discrete-time automatic pipeline, inventory and order-based production control system (APIOBPCS), a well-established production and inventory control model. The feedback mechanism within the replenishment rule enables the model to mitigate the bullwhip effect, but introduces a stability problem. In this research, a comprehensive stability analysis is conducted for arbitrary lead times using difference equation theory. On the basis of stability, a state space approach is advocated to analyse the impact of replenishment parameters, demand processes, and lead times on the robustness of the bullwhip effect. The stability results demonstrate that the production control system can easily be destabilised without incorporating the work-in-progress (WIP) feedback loop. Furthermore, it reveals that the stability problem for long lead times can be simplified with the stability condition independent of the lead time. The results obtained in this study provide useful guidelines for the selection of replenishment parameters to guarantee stability and mitigate the bullwhip effect.     

Multi-objective master production scheduling in make-to-order manufacturing

In this paper a multi-objective, long-term production scheduling in make-to-order manufacturing is considered and a lexicographic approach with a hierarchy of integer programming formulations is proposed. The problem objective is to allocate customer orders with various due dates among planning periods with limited capacities to minimize the number of tardy orders as a primary optimality criterion. Then, the maximum level of the input and output inventory is minimized as a secondary criterion, and finally the aggregate production is leveled over the planning horizon as an auxiliary criterion. A close relation between minimizing the maximal inventory and the maximum earliness of customer orders is shown and used to simplify the inventory leveling problem. Numerical examples, modeled after a real-world make-to-order flexible flowshop in a high-tech industry, are provided and some computational results are reported. The paper indicates that the maximum earliness of customer orders is an important managerial decision variable, and its minimum value can be applied to control the inventory of purchased materials and finished products to maximize the customer service level and minimize production costs.     

Aggregate production process design in global manufacturing using a real options approach

A main source of competitive advantage is derived from the cost efficiency offered by firms’ manufacturing and logistics operations. Consequently, firms typically globalise their operations whereby they may exploit the comparative advantages—defined as production functions—of the nations in which they are present. Production process design thus arises as a significant issue. The research presented in this paper targets two fundamental questions attached to production process design that multinational companies face, namely: (i) should plants that are located in different countries but producing similar products use similar production processes?; and (ii) given that the firm's policy is to use similar production processes, how should the production processes be designed? Among others, the paper shows, by way of a numerical illustration of a binational manufacturing network, that the option of choosing freely upon production process design for the respective facilities in certain cases adds little to firm value. In fact, the value of this option tends to zero as the volatility rate increases when the exchange rate is modelled as a geometric Brownian motion without drift rate, implying that firms should employ similar production processes throughout their manufacturing networks. That is, a market value approach stands up for the so-called copy-exactly approach to production process design in these settings. We furthermore show the effects of economies of scale on the optimal production process design.     

Congruence of manufacturing decision areas in a production system: a research framework

A well designed production system secures environmental and internal fit. Environmental fit in a production system refers to alignment of manufacturing decisions to the external settings such as product and market. Internal fit implies that manufacturing decisions are mutually supportive. This paper develops a framework to analyse congruence of manufacturing decision areas in a production system. The framework considers six broad manufacturing decision areas. Based on the literature review, 54 decision types and alternative decision choices for each decision type are identified. The subjective and/or objective constructs to measure decision type are presented which should be useful in designing construct and in data gathering to conduct empirical research. Using the proposed framework, many research questions concerning the settings of several decision types for a specific type of production system can be generated and empirically tested.     

Artificial neural networks as alternative tool for minimizing error predictions in manufacturing ultradeformable nanoliposome formulations

This work was aimed at determining the feasibility of artificial neural networks (ANN) by implementing backpropagation algorithms with default settings to generate better predictive models than multiple linear regression (MLR) analysis. The study was hypothesized on timolol-loaded liposomes. As tutorial data for ANN, causal factors were used, which were fed into the computer program. The number of training cycles has been identified in order to optimize the performance of the ANN. The optimization was performed by minimizing the error between the predicted and real response values in the training step. The results showed that training was stopped at 10?000 training cycles with 80% of the pattern values, because at this point the ANN generalizes better. Minimum validation error was achieved at 12 hidden neurons in a single layer. MLR has great prediction ability, with errors between predicted and real values lower than 1% in some of the parameters evaluated. Thus, the performance of this model was compared to that of the MLR using a factorial design. Optimal formulations were identified by minimizing the distance among measured and theoretical parameters, by estimating the prediction errors. Results indicate that the ANN shows much better predictive ability than the MLR model. These findings demonstrate the increased efficiency of the combination of ANN and design of experiments, compared to the conventional MLR modeling techniques.     

Application of Hybrid Learning Neural Fuzzy Systems in Reliability Prediction

This study presents a hybrid learning neural fuzzy system for accurately predicting system reliability. Neural fuzzy system learning with and without supervision has been successfully applied in control systems and pattern recognition problems. This investigation modifies the hybrid learning fuzzy systems to accept time series data and therefore examines the feasibility of reliability prediction. Two neural network systems are developed for solving different reliability prediction problems. Additionally, a scaled conjugate gradient learning method is applied to accelerate the training in the supervised learning phase. Several existing approaches, including feed‐forward multilayer perceptron (MLP) networks, radial basis function (RBF) neural networks and Box–Jenkins autoregressive integrated moving average (ARIMA) models, are used to compare the performance of the reliability prediction. The numerical results demonstrate that the neural fuzzy systems have higher prediction accuracy than the other methods. Copyright © 2005 John Wiley & Sons, Ltd.     

Handling uncertainties in modelling manufacturing processes with hybrid swarm intelligence

Seldom has research regarding manufacturing process modelling considered the two common types of uncertainties which are caused by randomness as in material properties and by fuzziness as in the inexact knowledge in manufacturing processes. Accuracies of process models can be downgraded if these uncertainties are ignored in the development of process models. In this paper, a hybrid swarm intelligence algorithm for developing process models which intends to achieve significant accuracies for manufacturing process modelling by addressing these two uncertainties is proposed. The hybrid swarm intelligence algorithm first applies the mechanism of particle swarm optimisation to generate structures of process models in polynomial forms, and then it applies the mechanism of fuzzy least square regression algorithm to determine fuzzy coefficients on polynomials so as to address the two uncertainties, fuzziness and randomness. Apart from addressing the two uncertainties, the common feature in manufacturing processes, nonlinearities between process parameters, which are not inevitable in manufacturing processes, can also be addressed. The effectiveness of the hybrid swarm algorithm is demonstrated by modelling of the solder paste dispensing process.     

Neural Network Burst Pressure Prediction in Composite Overwrapped Pressure Vessels Using Mathematically Modeled Acoustic Emission Failure Mechanism Data

The purpose of this research was to predict burst pressures in composite overwrapped pressure vessels (COPVs) by using mathematically modeled acoustic emission (AE) data. Both backpropagation neural network (BPNN) and multiple linear regression (MLR) analyses were performed on various subsets of the low proof pressure AE data to predict burst pressures and to determine if the two methods were comparable. AE data were collected during hydrostatic burst testing on the 15-inch diameter COPVs. Once collected, the AE data were filtered to eliminate noise then classified into AE failure mechanism data using a MATLAB Kohonen self-organizing map (SOM). The matrix cracking only amplitude distribution data were mathematically modeled using bounded Johnson distributions with the four Johnson distribution parameters – ?, λ, γ, and η – employed as inputs to make both the BPNN and MLR predictions. The burst pressure predictions generated using a MATLAB BPNN resulted in a worst case error of 1.997% as compared to ?1.666% for the MLR analysis, suggesting comparability. However, the MLR analysis required the data from all nine COPVs to get approximately the same results as the BPNN training on just five COPVs; plus, MLR analyses are intolerant to noise, whereas BPNNs are not.