Controlling non-conformities propagation in manufacturing

Aim: Recent mediatic recalls of products illustrate the losses in terms of cost and reputation induced by non-conformities reaching end customer. The purpose of this paper is to master the propagation of quality non-conformities in manufacturing processes. Most of the time, the control methods implemented to ensure the quality performance of a production system present weaknesses, inherent to their constitution. They can let non-conformities propagate along the value stream and even to the end customer inducing costly recalls for manufacturer. This paper presents a method to master non-conformities propagation in a production system. Method: In order to calibrate the proposition, two case studies have been issued on non-conformities propagation in a low-volumes industry in the energy field. Results: The proposed method combines two levels of analysis and associated improvement tools. At the event level, it proposes to build the control chart associated with the non-conformities propagation distance. At the system level, it proposes a propagation matrix to analyse the permeability of the global detection system and to identify where improvement effort has to be driven. Conclusion: The empirical results of the case studies indicate that the method is adapted in low-volume contexts.     

Level of compliance with the radiation protection regulation--a survey among Norwegian hospitals and X-ray institutes

To identify the level of compliance with the new radiation protection regulation among Norwegian health care enterprises (HCEs). Totally, 41 HCEs were authorised to use advanced X-ray equipment for medical purposes during 2005-07. Follow-up inspections with 14 HCEs were carried out during 2007-09. Main topics for the inspections were those requirements identified as most challenging to implement in the authorisation process. Totally, 192 non-conformities with the regulation were revealed during the authorisation process. The inspections revealed that 93 % of the inspected HCEs had non-conformities with the regulation. Most common non-conformities dealt with skills in radiation protection, establishment of local diagnostic reference levels, access to medical physicists and performance of quality control of X-ray equipment. Inspections are an effective tool for implementation of regulation the requirements at the HCEs, thus improving radiation protection awareness.     

Economic design of exponential charts for time between events monitoring

The occurrence of defects or non-conformities in manufacturing processes can usually be modeled by a homogeneous Poisson process. However, the process parameter may change over time and it can be monitored with statistical process control techniques. Control charts based on an exponential distribution, called exponential charts in this paper, can be developed to monitor the occurrence rate of such events. For manufacturers, the economic objective of production is very important and has to be optimized. An economic approach is developed in this paper for the design of exponential charts. We compare and contrast the performances of statistical design, economic design and economic–statistical design. The usefulness of the proposed economic design approach is justified. The relationships among these designs are illustrated through numerical examples. In particular, the economic–statistical design approach is interpreted from a multi-objective optimization viewpoint. The limitations of the approach as well as future research are also discussed.     

Integrated production planning and quality control for linear production systems under uncertainties of cycle time and finished product quality

This work considers serial production systems with several process steps and a possible quality control at final step. It deals with the problem of optimising planned lead time when the real lead time for each process is stochastic and the finished product quality is uncertain unless it is inspected. Three analytical models are proposed aiming to minimise the expected total cost, which is composed of the inventory and backlogging costs for the finished product and quality costs associated with inspection and non-conformities. These models correspond to three quality control policies: (i) without quality control, (ii) with quality control but without taking into account the inspection duration when optimising the planned lead time and (iii) with quality control and with considering the inspection duration when optimising the planned lead time. Based on the results, it can be highlighted the economic advantage of integrating quality control at the early stage of supply and production planning decisions for some cost parameters conditions. The robustness of the proposed models is also analysed regarding the variance of the probability distributions of the lead times.     

Estimating the time of step change with Poisson CUSUM and EWMA control charts

Knowing when a process has changed would simplify the search for and identification of the special cause. Consequently, having an estimate of the process change point following a control chart signal would be useful to process engineers. Much of the literature on change point models and techniques for statistical process control applications consider processes well modelled by the normal distribution. However, the Poisson distribution is commonly used in industrial quality control applications for modelling attribute-based process quality characteristics (e.g., counts of non-conformities). Some commonly used control charts for monitoring Poisson distributed data are the Poisson cumulative sum (CUSUM) and exponentially weighted moving average (EWMA) control charts. In this paper, we study the effect of changes in the design of the control chart on the performances of the change point estimators offered by these procedures. In particular, we compare root mean square error performances of the change point estimators offered by the Poisson CUSUM and EWMA control charts relative to that achieved by a maximum likelihood estimator for the process change point. Results indicate that the relative performance achieved by each change point estimator is a function of the corresponding control chart design. Relative mean index plots are provided to enable users of these control charts to choose a control chart design and change point estimator combination that will yield robust change point estimation performance across a range of potential change magnitudes.     

Process safety improvement--quality and target zero

Process safety practitioners have adopted quality management principles in design of process safety management systems with positive effect, yet achieving safety objectives sometimes remain a distant target. Companies regularly apply tools and methods which have roots in quality and productivity improvement. The "plan, do, check, act" improvement loop, statistical analysis of incidents (non-conformities), and performance trending popularized by Dr. Deming are now commonly used in the context of process safety. Significant advancements in HSE performance are reported after applying methods viewed as fundamental for quality management. In pursuit of continual process safety improvement, the paper examines various quality improvement methods, and explores how methods intended for product quality can be additionally applied to continual improvement of process safety. Methods such as Kaizen, Poke yoke, and TRIZ, while long established for quality improvement, are quite unfamiliar in the process safety arena. These methods are discussed for application in improving both process safety leadership and field work team performance. Practical ways to advance process safety, based on the methods, are given.     

SPC scheme to monitor linear predictors embedded in nonlinear profiles

Response Modeling Methodology (RMM) is a general platform to model monotone convex relationships. In this article, RMM is combined with linear regression analysis to model and estimate linear predictors (LPs) embedded in a nonlinear profile. A regression‐adjusted statistical process control scheme is then implemented to monitor the LP's residuals. To model and estimate the LP, RMM defines a Taylor series expansion of an unknown response transformation and then use canonical correlation analysis to estimate the LP. A possible hindrance to the implementation of the new scheme is possible occurrence of nonnormal errors (in violation of the linear regression model). Reasons for the occurrence of this phenomenon are explored and remedies offered. The effectiveness of the new scheme is demonstrated for data generated via Monte Carlo simulation. Results from hypothesis testing clearly indicate that the type of the response distribution, its skewness and the sample size, do not affect the effectiveness of the new approach. A detailed implementation routine is expounded, accompanied by a numerical example. When interest is solely focused on the stability of the LP, and the nonlinear profile per se is of little interest, the new general RMM‐based statistical process control scheme delivers an effective platform for process monitoring. Copyright © 2015 John Wiley & Sons, Ltd.     

Failure Analysis of a Cast Mn-Bronze Propeller

The investigation of a failure of a cast manganese bronze propeller, fractured after a short service period (~4–5 months), was conducted. Chemical analysis, visual inspection, optical microscopy, fractographic evaluation, SEM/EDS analysis, and hardness testing were used as the principal analytical tools for the failure investigation. The findings suggest that the failure occurred due to fatigue mechanism under low stress conditions. The non-conformities detected in material chemical composition, as well as the presence of internal and surface flaws could be considered as the main contributors of the premature failure of the propeller.     

非线性阻尼变刚度半主动结构振动控制

提出了具有非线性阻尼器的可变刚度系统及其半主动振动控制理论与方法，通过对５层钢框架试验模型的仿真计算，分析了非线性阻尼变化对结构振动响应的影响。分析结果表明：所提出的半主动控制系统能够有效地减小结构的振动反应。     

Modelling Frequency and Count Data

We present a general method for studying residual behavior of controlled autocorrelated processes subject to special-cause changes. Understanding residual response to a particular type of process change is important for selecting a proper control chart to monitor the process effectively. The method is introduced by analyzing the residual behavior of a controlled machining process described by a state-space model that has a simple structure and is representative of many real processes. The general method is then developed and applied to an ARIMA(p, l, q) process and to some special processes whose residual behavior has been investigated in the literature. Residual response to a general disturbance sequence is determined, and it is shown that the residual properties are independent of the type of feedback control applied to the process. Implications of this property for process control and monitoring are discussed.     

A Synthesis of Feedback and Feedforward Control for Process Improvement Under Stationary and Nonstationary Time Series Disturbance Models

Process adjustment strategy is an important part of the process improvement methods. The feedback control technique is used to compensate for the deviation of the output, and it has been intensively investigated. For continuous improvement and proactive strategies, feedback control has a delay and thus is not the ideal solution. In this article, motivated by a realistic manufacturing example, we propose the periodic shift disturbance models and investigate the feedforward control application from a new disturbance decomposition framework. We combine feedforward control with feedback control for maintaining the stability of the process and delivering products at target values. Then, we evaluate the performance of different control strategies for various disturbance models by using the output mean square error criterion. Sensitivity analysis of these control methods is made on different model parameter spaces, and robustness analysis for both model parameter and model structure misspecifications is presented. Two simulated examples show that the proposed control strategies can significantly reduce the variation of an evolving disturbance process. Copyright © 2014 John Wiley & Sons, Ltd.     

A WIP control policy for tandem lines

The control of manufacturing systems has many similarities with the management of inventories in warehouses. Based on the detailed analysis of these similarities, a new control policy is proposed in this paper. Simulation studies demonstrate that this new policy performs better than several established control policies by providing a higher service level with a lower work in process (WIP). This policy is in its infant stage of development and only single part type tandem lines (transfer lines) are studied in this paper. Studies of re-entrant production systems or multiple part type systems are the focus of future research.     

Multivariate statistical process control for autocorrelated processes

Multivariate statistical process control is often used in chemical and process industries where autocorrelation is most prevalent. We present a realistic model that generates autocorrelation and crosscorrelation and provides a useful approach to characterizing process data. We show how our model relates to the widely-used method of principal component analysis, distinguish between types of assignable causes, and present a useful control statistic based on a principal component decomposition that is not autocorrelated. The control chart for this statistic can be developed by a routine analysis even when the input data is autocorrelated. Furthermore, to characterize our results, we show that any linear combination of the input data that is not autocorrelated is related to our control statistic.     

汽车万向节叉热挤压成形的数值模拟分析

采用刚粘塑性有限元法对万向节叉热挤压成形过程进行数值模拟分析,综合考虑了变形、热传导、变形生热、摩擦生热等多个因素,得出了成形过程中金属流动变化的3个阶段。研究了摩擦条件和终锻温度对成形力的影响,得出良好的润滑条件和给模具进行预热,能够有效的控制成形力,有利于金属的流动及提高模具寿命。     

A Hybrid Learning‐Based Model for Simultaneous Monitoring of Process Mean and Variance

Unnatural patterns exhibited on process mean and variance control charts can be associated separately with different assignable causes. Quick and accurate knowledge of the type of control chart patterns (CCPs), either because of process mean or variance, can greatly facilitate identification of assignable causes. Over the past few decades, however, process mean and variance CCPs are seldom studied simultaneously in the statistical process control literature. This study proposes a hybrid learning‐based model for simultaneous monitoring of process mean and variance CCPs. In this model, a self‐organization map neural network‐based quantization error control chart is responsible for detecting the out‐of‐control signals, a discrete particle swarm optimization‐based selective ensemble of back‐propagation networks is responsible for classifying the detected out‐of‐control signals into categories of mean and/or variance abnormality, and two discrete particle swarm optimization‐based selective ensembles of learning vector quantization networks are responsible for further identifying the detected mean and variance out‐of‐control signals as one of the specific CCP types, respectively. Extensive simulations indicate that the proposed hybrid learning‐based model outperforms other existing approaches in detecting mean and variance changes, while also capable of CCP recognition. In addition, a case study is conducted to demonstrate how the proposed hybrid learning‐based model can function as an effective tool for monitoring mean and variance simultaneously. Copyright © 2013 John Wiley & Sons, Ltd.     

A broader view of the economic design of the X-bar chart in the semiconductor industry

This paper starts from a notice made in the semiconductor industry: a process control system and especially control charts provide information that can be exploited for correlation analyses during process investigations. In this industry, key and costly investigations are made for improving yield and reducing scrap. Daily, engineering teams are working at manufacturing improvements. Without process data, their work could take much more time and lead to weak improvements. Nevertheless, design of process control systems and in particular control charts lacks from taking into account this remark as there is no sound application to infer an optimal control chart depending on business parameters like yield, scrap, customer audits, etc. Meetings between several engineering teams (process control, quality, process integration, industrial engineering and production) occur frequently to find an affordable quantity of controls for each operation. The literature point of view does not provide more recommendations to take into account the reuse of data into these costly investigations. The paper investigates this issue. For this first investigation, work has been focused on the design economics of control charts for a simplified process model. The paper translates this concept into the Lorenzen and Vance's (1986) model. It simulates the design economic of a control chart taking into account this new model and infers new optimal statistical process control (SPC) set points. An analysis of this new link is made in a context of yield improvement, providing reference for knowing optimal quantity and frequency of controls.     

A Synthetic median control chart for monitoring the process mean with measurement errors

Recent studies show that the Shewhart median chart is widely used for detecting shifts in a process, but it is often rather inefficient in detecting small or moderate process shifts. In order to overcome this problem, a Synthetic chart can be used. This chart outperforms the Shewhart‐type chart because it uses the information about the time interval between two consecutive nonconforming samples. In this paper, we propose and study the Phase II Synthetic median control chart. A Markov chain methodology is used to evaluate the statistical performance of the proposed chart. Moreover, its performance is investigated in the presence of measurement errors, which are modelled by a linear covariate error model. We provide the results of an extensive numerical analysis with several tables and figures in order to show the statistical performance of the investigated chart, for both cases of measurement errors and no measurement errors. Finally, an example illustrates the use of the Synthetic median chart.     

An integrated EPQ model based on a control chart for an imperfect production process

The notion of quality control is introduced into the classic economic production quantity (EPQ) model. Based on previous research, this paper contributes to an integrated EPQ model combining the concepts of statistical process control and maintenance. An imperfect production process involving three different conditions is considered. A control chart is adopted to monitor the whole process with Taguchi's loss function estimating the quality cost of each condition. A corresponding maintenance policy is planned for the machine depending on what condition it runs in. Thus the proposed EPQ model takes quality-related costs and maintenance-related costs into account other than storage costs and ordering costs already considered in the classic model. The objective of this model is to minimise the total expected production cost per production cycle while simultaneously determining the optimal parameters of control chart design, the interval between sampling, the sample size and the maintenance decision policy. The pattern search method is used to solve the problem using the MATLAB toolbox. In addition, a case study, sensitivity analysis and comparison analysis are presented to demonstrate the application of the model.     

Applying Real-Time Statistical Process Control to Manufacturing Processes Exhibiting Between and Within Part Size Variability in the Wood Products Industry

In this paper, we develop a control charting procedure to monitor average part size, as well as between and within part size variation for sawn lumber in an automated lumbermill. We present a review of the sawing process followed by a discussion of sampling methods for a real-time noncontact laser measuring device. A statistical model based on the components of variation analysis of variance model is proposed both for the analysis of the data and the construction of control charts that can be used to monitor the process. The details of the resulting statistical process control system are developed and an example from the lumber industry is provided and compared to other possible approaches. The resulting techniques may have applicability in many other industries where within and between variation in processes occurs.