Analysis of Categorical Response in Small Sample Experiments

Statistical design of experiments (DOE) is widely used today for process and product characterization and optimization. Owing to cost and time considerations, sometimes only a minimum number of experimental runs can be conducted, with added challenges in analysis when the experimental outcomes cannot be measured on a continuous scale and are expressed only in qualitative terms such as ‘excellent’, ‘satisfactory’ and ‘poor’: such outcomes are variously described as ‘categorical’, ‘attribute’, ‘qualitative’, ‘discrete’ or ‘counted’ in nature. This paper offers practical techniques of handling small experiments with such non‐standard DOE response data which are otherwise impossible to analyze by standard statistical software. The suggested procedures, built upon what is called a Likelihood Transfer Function (LTF), do not require complex data analysis but would yield results consistent with the constraints of experimental conditions as well as the objectives of stakeholders. Copyright © 2015 John Wiley & Sons, Ltd.     

Event driven algorithms applied to a high energy ball mill simulation

In the physics community, event driven molecular dynamics simulations have been successfully applied to typical physical topics, such as granular gases. In contrast, in the engineering sciences, event driven algorithms have not become as popular, even though they can be far more efficient for special systems. In this paper, we present an event driven simulation of a high energy ball mill, where the motion and energy transfer of the grinding balls during operation are simulated. Stable numerical algorithms for collision detection and collision response are developed and several possible pitfalls are discussed. Furthermore, an improved event list handling technique and a specialized space subdivision method is presented. The performance of the partitioning is demonstrated by experiments. It is shown that the scaling rules usually applied are oversimplified. A new way of scaling the process parameters to obtain a higher production yield using iso-impact diagrams is presented.     

Subject-orientation as design language for integration across organisational control layers

Recent research and development in production industry reveals a movement to cyber-physical production systems and Internet of things enabled manufacturing. In this context, ways in which enterprise processes are conceptualised and executed are changing. Decentralising production by applying interlinked cyber-physical production resources breaks up the traditional boundaries between different process abstraction layers. Heterogeneous smart devices and processes at all levels in the industrial control need to interact in such systems. In this paper, requirements from the field of Business Process Management towards the vertical integration of business and production processes are derived. With respect to the identified requirements the eligibility of the Subject-oriented Process Management (S-BPM) approach for vertical process integration is depicted. At the core, the paper investigates the applicability of S-BPM to achieve vertical process integration and reports lessons learnt from two industrial application scenarios. The first scenario aims to encode human properties to adapt workplaces to human needs. The second scenario aims towards supporting the execution and tracking of dynamic production processes.     

Optimization of Multiple Response Surfaces with Secondary Constraints for Improving a Radiography Inspection Process

The quality and reliability measures of products and processes are frequently multidimensional, and the goal of various statistical studies and experiments is to find settings of design or process variables that optimize such products or processes. In many such situations, these measures of interest are further restricted by some physical secondary constraints of the process. These constraints may cause interdependencies among the factors. Moreover, the operability region imposed by the constraints is often unknown a priori, and hence, standard experimental strategy is not always feasible. In this paper, we study a radiography inspection process with dual conflicting radiograph quality measures, contrast sensitivity and spatial resolution, and an image density constraint. The paper presents a sequential approach to tackle this constrained optimization problem. We use and initial experiment to estimate the constraints, and then to account for the interdependencies between the factors more accurately, we use a sliding level design with sliding factors nested within multiple factors. We propose to use a nested effects modeling approach to analyze the experimental data and illustrate its benefits for this example. To optimize conflicting multiple responses, a goal optimization formulation based on this modeling approach is presented.     

Evolutionary experiments for self-assembling amphiphilic systems

Some amphiphilic molecules in particular environments may self-assemble and originate chemical entities, such as vesicles, which are relevant in technological applications. Experimentation in this field is difficult because of the high dimensionality of the search space and the high cost of each experiment. To tackle the problem of designing a relatively small number of experiments to achieve the relevant information on the problem, we propose an evolutionary design of experiments based on a genetic algorithm. We built a particular algorithm where design and laboratory experimentation interact leading the search toward the optimality region of the space. To get insight in the process we then modelled the experimental results with different classes of regression models; from modelling we could identify the special role played by some molecules and the relevance of their relative weight in the composition. With modelling we “virtually” explored the experimental space and predicted compositions likely to generate very high yields. Models then provide valuable information for the redesign of the experiments and can be considered as an essential addition to the evolutionary approach.     

Design of experiments for multi-stage processes

Many industrial processes involve large numbers of input variables so that a complete factorial design would require prohibitively many experimental runs. This is particularly the case with multi-stage processes. We offer a view of multi-stage processes, in terms of autoregressive models, which can help in the sensible reduction in the size of experiments and can also allow different stages of the process to be investigated separately.     

Molecular dynamics study of the growth of optical fibers

We have carried out molecular dynamics simulations that model a special growth technique for optical fibers, the so-called laser heated pedestal growth. Although the simulations are a crude representation of the experimental process, they demonstrate that it is possible to simulate such experiments on a microscopic scale. We measure properties of both the growth process and the resulting fibers and find some of those properties to match our intuitive expectations while others show a more complicated behavior.     

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.     

Optimal designs for dual response systems for the normal and binomial case

Most research in design of experiments focuses on appropriate designs for a system with just one type of response, rather than multiple responses. In a decision-making process, relying on only one objective can lead to oversimplified, suboptimal choices that ignore important considerations. Consequently, the problem of constructing a design for an experiment when multiple types of responses are of interest often does not have a single definitive answer, particularly when the response variables have different distributions. Each of these response distributions imposes different requirements on the experimental design. Computer-generated optimal designs are popular design choices for less standard scenarios where classical designs are not ideal. This work presents a new approach to experimental designs for dual-response systems. The normal and binomial distributions are considered as potential responses. Using the D-criterion for the linear model and the Bayesian D-criterion for the logistic regression model, a weighted criterion is implemented in a coordinate-exchange algorithm. Designs are evaluated and compared across different weights. The sensitivity of the designs to the priors supplied for the Bayesian D-criterion is also explored.     

基于包容性设计流程视角下的助行器开发研究

目的 深入探讨包容性设计作为一种以人为本的设计方法和流程,而不仅仅是设计理念,在残障人士康复辅具开发流程中的运用价值。方法 结合复健拐杖的设计实例,运用剑桥工程中心开发的“设计轮”,将包容性设计流程四个阶段17个步骤结合用户共创设计方法,通过焦点小组、用户画像、用户旅程图等方式列出需求清单,运用形态分析法和TRIZ创新法进行设计创意,快速制作原型后与参与用户和设计师共同测试、深化方案,并进一步对设计排除进行评估后定案。结论 通过包容性设计理论结合助行器开发流程,探讨了包容性设计方法与多种创意方法融合的可行性,为更多康复辅具的包容性设计创新提供了实操性的参考思路和可行性案例,使包容性设计不仅停留在理念层面,还落实在具体设计操作实践流程上。     

An information management paradigm for statistical experiments

Optimization of engineering processes and products via statistical design of experiments is an approach well known to statisticians but still not popularly used by technical personnel. This paper sets out a pattern of reasoning that would facilitate the appreciation, on the part of non‐mathematicians, of the principles and advantages of using statistical experimental design for process and product modeling and optimization. Use is made of the concepts of information transformation and conservation in a language that is familiar to those of purely technical background, leading to better understanding, acceptance and application of the efficiency and effectiveness of statistical experiments. In today's environment of the prevalence of software and hardware for statistical analysis, engineers concerned with quality and reliability would particularly benefit from such a paradigm for process and product performance improvement. Copyright © 2009 John Wiley & Sons, Ltd.     

Better integration of human factors considerations within safety in design

‘Safety in design’ is becoming an increasingly used approach in a variety of domains. Also known as ‘safe design’ or ‘prevention through design’, the general process seeks to eliminate health and safety hazards, or minimise potential risks, by systematically involving end-users and decision makers in the full life cycle of the designed product or system. This article focuses on the need to better integrate human factors considerations within safe design for high-hazard industries. In particular, it describes the benefits of applying a task-based approach to assist design processes. By way of presenting a safe design method for equipment used in mining, it shows the need to obtain end-user input both to better reveal design deficiencies and to identify effective solutions for future equipment. Overall, a multidisciplinary framework is advocated, in particular by blending human factors techniques and structured risk management methods with the existing equipment design processes.     

Framework on robotic percussive riveting for aircraft assembly automation

Presented in this paper is a framework for the implementation of a robotic percussive riveting system, a new robot application for aircraft assembly. It is shown here that a successful robot application to the automation of a process requires in-depth research of the process and the interaction with the robot. For this purpose, a process planning-driven approach is proposed to guide a robot application research.Atypical process planning will involve a list of
key considerations including: process sequence, process parameters, process tooling, and process control. Through this list, a number of key research issues are identified for robotic percussive riveting, such as rivet pattern planning,rivet time determination, rivet tooling design and rivet insertion control. The detailed research on these issues has effectively created know-how for the successful implementation of our robotic percussive riveting system.     

Exploring the Use of Design of Experiments in Industrial Processes Operating Under Closed‐Loop Control

Industrial manufacturing processes often operate under closed‐loop control, where automation aims to keep important process variables at their set‐points. In process industries such as pulp, paper, chemical and steel plants, it is often hard to find production processes operating in open loop. Instead, closed‐loop control systems will actively attempt to minimize the impact of process disturbances. However, we argue that an implicit assumption in most experimental investigations is that the studied system is open loop, allowing the experimental factors to freely affect the important system responses. This scenario is typically not found in process industries. The purpose of this article is therefore to explore issues of experimental design and analysis in processes operating under closed‐loop control and to illustrate how Design of Experiments can help in improving and optimizing such processes. The Tennessee Eastman challenge process simulator is used as a test‐bed to highlight two experimental scenarios. The first scenario explores the impact of experimental factors that may be considered as disturbances in the closed‐loop system. The second scenario exemplifies a screening design using the set‐points of controllers as experimental factors. We provide examples of how to analyze the two scenarios. © 2017 The Authors Quality and Reliability Engineering International Published by John Wiley & Sons Ltd     

Real-time production control of an FMS in a produce-to-order environment

This paper deals with real-time control of an FMS, operating in a produce-to-order environment, with machines subject to failure. A two-level production control system (PCS) is developed and examined. The proposed PCS incorporates three principles: no need for pre-planning procedures; separation between due-date and operational considerations; and full exploitation of process flexibility. In an extensive series of simulation experiments it was demonstrated that the proposed PCS leads to good results in both criteria: maximum throughput and minimum tardiness of orders. A comparison study between this PCS and one which is not based on separation between due-date and operational considerations, indicated the advantage of the former.     

Selection of a pull production control system in multi-stage production processes

In this paper, we conduct an analytical comparison of three pull production control systems: Kanban, CONWIP and Base-stock in multi-stage production processes. First, we compare the three control systems in a multi-stage serial production process. Then, we compare them in multi-stage assembly production processes, and present guidelines that allow us to select the best system. As a result, we show which structural parameters decide the superiority of one control scheme to the others, and how they are related. A key for superiority is a configuration of parameters, such as processing times and number of cards employed in the system. We show that there is no general superiority amongst the analysed concepts. Finally, we verify the effect of variability on the system performance, and generalise the analytical results of deterministic cases by conducting numerical experiments.     

A note on GLR charts for monitoring count processes

Various generalized likelihood ratio (GLR) charts have been proposed to monitor count processes such as binomial, Bernoulli, Poisson, and multinomial processes. The advantages of GLR charts are that designing the chart is relatively easy, estimates of the process change‐point and shift size are available for post‐signal diagnosis, and they are effective in detecting a wide range of shifts in the process parameter. However, for some special cases of the observations, such as observing all defective items or all non‐defective items, the GLR chart statistic for monitoring a count process has been said to be undefined. We show that the GLR chart statistic is always well defined.     

Towards Improved Analysis Methods for Two‐Level Factorial Experiments with Time Series Responses

Dynamic processes exhibit a time delay between the disturbances and the resulting process response. Therefore, one has to acknowledge process dynamics, such as transition times, when planning and analyzing experiments in dynamic processes. In this article, we explore, discuss, and compare different methods to estimate location effects for two‐level factorial experiments where the responses are represented by time series. Particularly, we outline the use of intervention‐noise modeling to estimate the effects and to compare this method by using the averages of the response observations in each run as the single response. The comparisons are made by simulated experiments using a dynamic continuous process model. The results show that the effect estimates for the different analysis methods are similar. Using the average of the response in each run, but removing the transition time, is found to be a competitive, robust, and straightforward method, whereas intervention‐noise models are found to be more comprehensive, render slightly fewer spurious effects, find more of the active effects for unreplicated experiments and provide the possibility to model effect dynamics. Copyright © 2012 John Wiley & Sons, Ltd.     

Interactive data mining informs designed experiments

We illustrate how a Six Sigma project team can apply recursive partitioning to a historical data set to narrow down a list of potential experimental factors and then construct an experimental design using information from the partition analysis. The paper illustrates the value of analyzing historical manufacturing data to inform the choice of factors and levels for statistically designed experiments. Copyright © 2008 John Wiley & Sons, Ltd.     

Engineering concerns and new developments in anaerobic waste-water treatment

A non-comprehensive review of several developments in the field of anaerobic biological waste-water treatment engineering is carried out, considering the active role engineers have to play in this field. The analysis is done as a result of the fact that the general performance of anaerobic systems and the diversity of wastes that these can treat are increasing as a result of new reactor design, operating conditions, and use of specialized microorganisms. This paper illustrates a few examples of conventional and new applications of anaerobic treatment and several advantages, such as no energy-intensive oxygen transfer, less excess sludge, production of combustible biogas, and space saving. Such an overview of biological waste-water treatment also precedes comments on some important aspects on process microbiology as well as considerations of the application of fundamentals and kinetics to the analysis of the biological processes used most commonly for anaerobic biological waste-water treatment. Models developed for this reaction type are discussed, considering four overall steps. Also, models for a continuous-stirred tank bioreactor, a fluidized bed reactor, and a packed bed reactor as basic configurations are described. Major anaerobic biological treatment processes used for waste-water treatment, both suspended-growth processes and reactors (anaerobic stirred tank, anaerobic contact process, upflow anaerobic sludge-blanket process) and attached-growth processes and reactors (anaerobic filter process, expanded-bed process), are considered. It was also shown that the breakthroughs dealing with reactor design and operating conditions offer practical solutions to many of the drawbacks that were initially thought to limit the purpose of anaerobic processes. Finally, the paper includes some aspects regarding the control of anaerobic biological systems. Electronic Publication