A multivariate dataset consists of n cases in d dimensions, and is often stored in an n by d data matrix. It is well-known that real data may contain outliers. Depending on the situation, outliers may be (a) undesirable errors which can adversely affect the data analysis, or (b) valuable nuggets of unexpected information. In statistics and data analysis the word outlier usually refers to a row of the data matrix, and the methods to detect such outliers only work when at least half the rows are clean. But often many rows have a few contaminated cell values, which may not be visible by looking at each variable (column) separately. We propose the first method to detect deviating data cells in a multivariate sample which takes the correlations between the variables into account. It has no restriction on the number of clean rows, and can deal with high dimensions. Other advantages are that it provides predicted values of the outlying cells, while imputing missing values at the same time. We illustrate the method on several real data sets, where it uncovers more structure than found by purely columnwise methods or purely rowwise methods. The proposed method can help to diagnose why a certain row is outlying, e.g. in process control. It also serves as an initial step for estimating multivariate location and scatter matrices.

Multivariate data are typically represented by a rectangular matrix (table) in which the rows are the objects (cases) and the columns are the variables (measurements). When there are many variables one often reduces the dimension by principal component analysis (PCA), which in its basic form is not robust to outliers. Much research has focused on handling rowwise outliers, i.e. rows that deviate from the majority of the rows in the data (for instance, they might belong to a different population). In recent years also cellwise outliers are receiving attention. These are suspicious cells (entries) that can occur anywhere in the table. Even a relatively small proportion of outlying cells can contaminate over half the rows, which causes rowwise robust methods to break down. In this paper a new PCA method is constructed which combines the strengths of two existing robust methods in order to be robust against both cellwise and rowwise outliers. At the same time, the algorithm can cope with missing values. As of yet it is the only PCA method that can deal with all three problems simultaneously. Its name MacroPCA stands for PCA allowing for Missingness And Cellwise & Rowwise Outliers. Several 1 arXiv:1806.00954v3 [stat.ME] 9 Dec 2018 simulations and real data sets illustrate its robustness. New residual maps are introduced, which help to determine which variables are responsible for the outlying behavior. The method is well-suited for online process control.Supplementary material is available online.

Increasing resource prices, ever-higher complexity of products, recent developments in legislation and the importance of a green brand image have resulted in an increased interest of original equipment manufacturers to facilitate a disassembly based end-of-life treatment for their products. The main reason is that precious metals, rare earth elements and plastics can be recovered with the highest yield and purity in a disassembly based treatment. However, original equipment manufacturers currently face several issues for the implementation of design for disassembly. To overcome these issues, first of all an in-depth analysis of design-for-disassembly opportunities and challenges is presented. Taking into account the results of this analysis, innovative low-cost elastomer-based fasteners have been developed, which can be simultaneously released by applying a sufficiently high force over a period of time. In addition, an experimental validation method was developed and adopted to demonstrate that the developed fasteners allow reducing the disassembly time by 70% to 90% for the housing of LCD TVs without compromising product robustness. The presented calculations indicate that the implementation of the developed fasteners is profitable from an overall perspective in regions with a labor cost higher than 7 €/h. However, original equipment manufacturers currently lack incentives to adopt design for disassembly for products sold in a Business-to-Consumer market, which are jointly collected and treated at end-of-life. Therefore, a differentiation in recycling fees proportional to the reduction in disassembly time is proposed to provide economic stimuli for original equipment manufacturers to implement these fasteners. Such a differentiation scheme, combined with the presented insights on opportunities to facilitate disassembly processes and the required resistance of fasteners to forces in function of time, will stimulate and enable the development of products which can be disassembled in an economically viable manner, resulting in improved material recovery from end-of-life products in industrialized regions.

This paper presents a number of novel active fasteners developed to significantly lower disassembly costs during reconditioning, remanufacturing, and recycling of products. In the initial stage of the fastener development process, the applicability of distinct trigger signals for active disassembly (AD) is evaluated. Based on this evaluation, the high robustness of using a pressure increase or decrease as a nondestructive trigger for AD is demonstrated. Since previously proposed pressure-sensitive fasteners face considerable drawbacks upon implementation in electronic products due to the ongoing trend of miniaturization, a second generation of pressure-based active fasteners is developed. Evaluation of these fasteners by means of axiomatic design techniques and prototyping demonstrates that the presented snap-fits, which make use of a closed-cell elastomer foam, are most robust. Subsequently, the contraction forces that closed-celled foams can exert as a function of an increase in ambient air pressure are experimentally determined. Furthermore, the implementation of pressure-sensitive foam-based snap-fits in both a modem and a payment terminal is described. Results of these experiments demonstrate that the contraction force of a cross-linked metallocene polyethylene closed-cell foams can reach up to 6 N/cm 2 at an overpressure of 2 bar and that the foam-based snap-fits can be released at a pressure increase of 2 bar.

No abstract