A study on quantitative analysis of human reliability with inspection

This paper is concerned with determining human reliability with inspection. The two-state models of operation-inspection sequence with certain dependence relations are shown where all working units are supposed to have two states: success or failure. Moreover, the multi-state models of operation-inspection are presented on the assumption that operation unit is of multi-state and each inspection unit is in two states.     

Carryover effects in multi-attribute inspection

Consider a product that has several quality attributes for inspection where the inspection of some attributes may have carryover effect(s) on subsequent inspections of other attributes. In this case, not all the attributes can be inspected on the same unit and therefore more than one unit may be required to complete one inspection cycle of all the attributes. Furthermore, if the inspection time of each attribute is considered, the number of attributes and their inspection sequence on each inspected unit may affect the time-span of an inspection cycle. In this paper, heuristic procedures are developed to find the inspection sequences of all the attributes that minimize the total number of inspected units and to balance the inspection time of each inspected unit such that the time-span of an inspection cycle is minimized. An integrated circuit card inspection example is given to illustrate the implementation of these procedures.     

Automated classification of electrical appliances on a production line

User simulation inspection of every appliance on a production line is time consuming and expensive. A more effective way is to use sensors for fast indirect measurements of selected quality indexes, ie appliance features, which may be used for automatic on-line inspection and classification, by correlation to manual inspection. Feature vectors for classifying electrical appliances tend to form overlapping, irregular amorphous clusters in a multidimensional feature space. Three classifier algorithms were formulated to address this difficult classification problem, which is aggravated by the requirement that almost no bad units should be misclassified as good ones. The discriminatory power of two or all three classifiers is combined by a classifier voting strategy. The different classifiers and voting strategies are compared in terms of four performance indexes, predicting the percentage of bad units sent to the customer, percentage of good units rejected as bad ones, a cost-weighted class contamination index and the expected percentage of correctly classified units. Practical application is implemented on a feature data base of several hundred labelled refrigerators, whereby it is demonstrated that three classifier c voting will practically never misclassify a bad unit as a good one.     

An optimal inspection strategy for randomly failing equipment

This paper addresses the problem of generating optimal inspection strategies for randomly failing equipment where imminent failure is not obvious and can only be detected through inspection. Inspections are carried out following a condition-based procedure. The equipment is replaced if it has failed or if it shows imminent signs of failure. The latter state is indicated by measuring certain predetermined control parameters during inspection. Costs are associated with inspection, idle time and preventive or corrective actions. An optimal inspection strategy is defined as the inspection sequence minimizing the expected total cost per time unit over an infinite span. A mathematical model and a numerical algorithm are developed to generate an optimal inspection sequence. As a practical example, the model is applied to provide a machine tool operator with a time sequence for inspecting the cutting tool. The tool life time distribution and the trend of one control parameter defining its actual condition are supposed to be known.     

Sequential inspection for SPC-quarantined production

The problem of inspecting process output in response to a control chart out-of-control signal is addressed. A Bayesian sequential inspection procedure is developed for the case where units can be inspected in reverse of the order in which they were produced. The inspection procedure is used to determine whether the 'out-of-control' production is of sufficiently poor quality to merit further inspection, while simultaneously attempting to curtail inspection by identifying the point at which the process went out-of-control. Examples are presented, and Monte Carlo analysis is used to demonstrate the performance characteristics of the inspection procedure.     

Weibull prediction of a future number of failures

This paper presents simple new prediction limits for the number of failures that will be observed in a future inspection of a sample of units. The past data consist of the cumulative number of failures in a previous inspection of the same sample of units. Life of such units is modelled with a Weibull distribution with a given shape parameter value. Copyright © 2000 John Wiley & Sons, Ltd.     

Über die Fehlerauffindbarkeit als Hauptforderung an die zerstörungsfreie Werkstoffprüfung

About the Detection of Imperfections as a Main Task for Nondestructive Testing The inspection uncertainty is identified which is a measure of the difference between the size of imperfection that has a high probability of being passed and the size of imperfection that has a high probability of being rejected by an inspection. Reduction of the inspection uncertainty is the only means by which one can reduce inspection errors as the acceptance of defective material units and the rejection of sound material units. The discussion is presented within a framework that allows a quantitative evaluation of the quality of the material units that pass an inspection and a determin ation of the rejection rate. Analysis of the causes of inspection uncertainty is essential to efficient strategies for improving nondestructive inspection capabilities.     

Bayesian framework for prediction of future number of failures from a single group of units in the field

This paper considers prediction of unknown number of failures in a future inspection of a group of in-service units based on number of failures observed from an earlier inspection. We develop a flexible Bayesian model and calculate Bayesian estimator for this unknown number and other quantities of interest. The paper also includes an illustration of our method in an example about heat exchanger. A main advantage of our approach is in its nonparametric nature. By nonparametric here we simply mean that no assumption is required about the failure time distribution of a unit.     

Development of methods and means of coordinate measurements for linear and angular parameters of cutting instruments

A method is proposed for contact and contactless coordinate measurements of the geometrical parameters of cutting instruments. Practical applications of coordinate measuring instruments and systems for the inspection of cutting tools and the setup of tool units are discussed.     

Inspection strategies for multistage production systems with time-varying quality

An economic, cost of quality model is formulated for a production system with multiple inspection locations and sampling. The model is applied to a thermal printer for digital photographs to determine when inspection of incoming material is appropriate. A simple dynamic model of quality is used, allowing for common-cause defects and the evaluation of sampling plans. The opportunity to correct quality problems more quickly when they are detected earlier is also considered. The cost structure used allows the costs of quality improvement, and not just that of defective units, to be considered.     

Viewing Strategy and Window Effects on Dynamic Visual-Inspection Accuracy

Equally-spaced product units, moving on a belt at a uniform velocity, were exposed for visual inspection over a fixed viewing window. Viewing windows were equal to the product unit interspacing and twice this distance to give respectively 1 and 2 product units within the window. Three different belt speeds were used. Subjects performed short runs of visual inspection using their self-directed strategies of viewing the sequence of product units and then these subjects repeated all conditions of this task using a “directed viewing strategy.” Improved viewing discipline was expected from this “directed” strategy and eye motion monitoring was employed to verify that the subjects used the “directed” strategy. Inspection errors of each type were observed under twelve experimental conditions. Both types of errors were found to be greater with: 1) the smaller viewing window, 2) self-directed viewing strategies in contrast to the “directed” strategy, and 3) higher belt velocities. Contrasts were also made between error types and between high- and low-performing subject groups. Inferences on the apparent cost of each error type were made from these data and compared with rational objectives. Economic implications are shown for these viewing-strategy and viewing window effects.     

The Maximum Likelihood Estimation Under Curtailed Three Class Attributes Plans

In this paper Semi-curtailed and Fully-curtailed three class attributes plans are introduced. The maximum likelihood estimators of the proportion of bad units and that of marginal units under these plans are obtained. The efficiency of these estimators in terms of saving in inspection is discussed. A graphical procedure to simplify the execution of these plans is provided.     

Economic optimization of off-line inspection in a process subject to failure and recovery

In certain types of processes, verification of the quality of the output units is possible only after the entire batch has been processed. We develop a model that prescribes which units should be inspected and how the units that were not inspected should be disposed of, in order to minimize the expected sum of inspection costs and disposition error costs, for processes that are subject to random failure and recovery. The model is based on a dynamic programming algorithm that has a low computational complexity. The study also includes a sensitivity analysis under a variety of cost and probability scenarios, supplemented by an analysis of the smallest batch that requires inspection, the expected number of inspections, and the performance of an easy to implement heuristic.     

A control chart for monitoring process mean based on attribute inspection

This article proposes a new control chart, namely the MON chart, which employs attribute inspection (inspecting whether units are conforming or nonconforming) to monitor the mean value of a variable characteristic x. A unit is classified as nonconforming if the value of x falls beyond a fixed warning limit. A sample is regarded as suspect if more than m out of n units (referred to as MON) in the sample are nonconforming. A MON chart produces an out-of-control signal when the interval between two suspect samples is smaller than a control limit. The MON chart is distinctively advantageous owing to its simplicity in implementation. In particular, the MON chart uses attribute inspection and eliminates the need for any computation. In addition, the MON chart makes use of information not only about the magnitude of x, but also the interval between two suspect samples. Therefore, it always outperforms the X chart and often excels the CUSUM chart on the basis of same inspection cost. Furthermore, the MON chart performs more uniformly over a wide range of mean shift than other charts.     

Economic optimization of off-line inspection

In this paper we solve the problem of determining the optimal inspection/disposition policy for a finite batch of items produced by a machine that is subject to random breakdowns. In particular, we identify which units should be inspected and in which order so as to minimize costs. The operational implications of the optimal policy are analyzed with a selected set of numerical results. We place special emphasis on three different policies: the cost minimizing policy; the policy of perfect information, i.e., we insist on determining the quality of each unit; and the policy of zero-defects,i.e., we insist that all accepted units are known to conform to specifications, allowing the rejection of units of unknown quality. We also show how the optimal inspection/disposition policy is incorporated into the optimization of the batch size.     

Inspector location in convergent production lines

This paper develops a finite-horizon denumerable state-space dynamic programming formulation of the inspector location problem in convergent production lines. The effects of the inspection point location decisions on production planning are clarified, and the resulting complexity of the determination of the costs of not detecting defective units is demonstrated under the assumption that all detected defective units are eliminated from the production line. A sample problem is solved by a backward induction algorithm to obtain the minimum-cost screening program.     

Inspection Error Adjustment in the Design of Single Sampling Attributes Plan

This article presents an approach for the design of single sampling attributes plans of given strength when inspection errors are present. The design criterion considered is the specification of two points of the Operating Characteristic curve. The design approach also recommends repeated testing of sampled nonconforming units for conformance in order to reduce the adverse effect of the inspection errors. It is shown that realistic sampling plan parameters can be obtained if the inspection error probabilities are taken into account in the design of sampling plans.     

Economic design of offline inspections for a batch production process

We consider a batch production process that can be either stable or unstable, in which inspection is performed offline after production of the batch is completed. The quality of a batch can be estimated with a desired level of certainty by inspecting only a sample of its units. In order to minimise the expected total cost per batch, which includes the costs of inspection, of false acceptance and of false rejection, we propose an economic inspection plan in which only a fraction of the batches, rather than each batch, is inspected. We prove that the expected total cost is a strictly quasiconvex function of the inspection interval. We establish necessary and sufficient conditions for the optimal inspection interval to be finite and propose an efficient algorithm to obtain its value. We demonstrate for the case of a single-sampling plan where the proposed economic approach outperforms the common procedure of inspecting every batch, and the proposed algorithm is very efficient.     

A Cost-Minimization Model for Multicharacteristic Component Inspection

In this paper a mathematical model is developed for determining the optimal number of inspections for multicharacteristic components where failure can be catastrophic. The model optimizes the expected total cost per accepted component resulting from (1) type I errors, (2) type II errors, (3) cost of added inspection, and (4) ordering of quality characteristics for inspection. The model considers components with several characteristics to be inspected. Failure to meet the quality requirements of any one characteristic results in the rejection of the component. Taking into consideration all three costs referred to above, a mathematical expression for expected total cost per accepted component is obtained. Also, an expression is developed for finding the optimal sequence of characteristic inspection. Finally, a computational procedure is outlined to determine the optimal sequence of characteristic inspection and the optimal number of inspections using the two expressions stated above.     

Dynamic quality control in assembly systems

We study an assembly system, in which each end product consists of two different components. The components are produced in batches, with possible defective units. The proportion of defective units in each batch is itself a random variable, known only in terms of its distribution. Defective units can be identified and corrected through an inspection-repair procedure, before final assembly taking place. We are concerned with developing an optimal inspection policy that minimizes the total expected cost due to inspection, repair, and warranty. Particularly, we focus on policies that have simple threshold structures and a near-optimal performance.