Clustering and visualization of failure modes using an evolving tree

Despite the popularity of Failure Mode and Effect Analysis (FMEA) in a wide range of industries, two well-known shortcomings are the complexity of the FMEA worksheet and its intricacy of use. To the best of our knowledge, the use of computation techniques for solving the aforementioned shortcomings is limited. As such, the idea of clustering and visualization pertaining to the failure modes in FMEA is proposed in this paper. A neural network visualization model with an incremental learning feature, i.e., the evolving tree (ETree), is adopted to allow the failure modes in FMEA to be clustered and visualized as a tree structure. In addition, the ideas of risk interval and risk ordering for different groups of failure modes are proposed to allow the failure modes to be ordered, analyzed, and evaluated in groups. The main advantages of the proposed method lie in its ability to transform failure modes in a complex FMEA worksheet to a tree structure for better visualization, while maintaining the risk evaluation and ordering features. It can be applied to the conventional FMEA methodology without requiring additional information or data. A real world case study in the edible bird nest industry in Sarawak (Borneo Island) is used to evaluate the usefulness of the proposed method. The experiments show that the failure modes in FMEA can be effectively visualized through the tree structure. A discussion with FMEA users engaged in the case study indicates that such visualization is helpful in comprehending and analyzing the respective failure modes, as compared with those in an FMEA table. The resulting tree structure, together with risk interval and risk ordering, provides a quick and easily understandable framework to elucidate important information from complex FMEA forms; therefore facilitating the decision-making tasks by FMEA users. The significance of this study is twofold, viz., the use of a computational visualization approach to tackling two well-known shortcomings of FMEA; and the use of ETree as an effective neural network learning paradigm to facilitate FMEA implementations. These findings aim to spearhead the potential adoption of FMEA as a useful and usable risk evaluation and management tool by the wider community.

Fat mass- and obesity-associated genotype, dietary intakes and anthropometric measures in European adults: the Food4Me study

The interplay between the fat mass- and obesity-associated (FTO) gene variants and diet has been implicated in the development of obesity. The aim of the present analysis was to investigate associations between FTO genotype, dietary intakes and anthropometrics among European adults. Participants in the Food4Me randomised controlled trial were genotyped for FTO genotype (rs9939609) and their dietary intakes, and diet quality scores (Healthy Eating Index and PREDIMED-based Mediterranean diet score) were estimated from FFQ. Relationships between FTO genotype, diet and anthropometrics (weight, waist circumference (WC) and BMI) were evaluated at baseline. European adults with the FTO risk genotype had greater WC (AA v. TT: +1·4 cm; P=0·003) and BMI (+0·9 kg/m2; P=0·001) than individuals with no risk alleles. Subjects with the lowest fried food consumption and two copies of the FTO risk variant had on average 1·4 kg/m2 greater BMI (Ptrend=0·028) and 3·1 cm greater WC (Ptrend=0·045) compared with individuals with no copies of the risk allele and with the lowest fried food consumption. However, there was no evidence of interactions between FTO genotype and dietary intakes on BMI and WC, and thus further research is required to confirm or refute these findings.

How reliable is internet-based self-reported identity, socio-demographic and obesity measures in European adults?

In e-health intervention studies, there are concerns about the reliability of internet-based, self-reported (SR) data and about the potential for identity fraud. This study introduced and tested a novel procedure for assessing the validity of internet-based, SR identity and validated anthropometric and demographic data via measurements performed face-to-face in a validation study (VS). Participants (n = 140) from seven European countries, participating in the Food4Me intervention study which aimed to test the efficacy of personalised nutrition approaches delivered via the internet, were invited to take part in the VS. Participants visited a research centre in each country within 2 weeks of providing SR data via the internet. Participants received detailed instructions on how to perform each measurement. Individual's identity was checked visually and by repeated collection and analysis of buccal cell DNA for 33 genetic variants. Validation of identity using genomic information showed perfect concordance between SR and VS. Similar results were found for demographic data (age and sex verification). We observed strong intra-class correlation coefficients between SR and VS for anthropometric data (height 0.990, weight 0.994 and BMI 0.983). However, internet-based SR weight was under-reported (Δ -0.70 kg [-3.6 to 2.1], p < 0.0001) and, therefore, BMI was lower for SR data (Δ -0.29 kg m(-2) [-1.5 to 1.0], p < 0.0001). BMI classification was correct in 93 % of cases. We demonstrate the utility of genotype information for detection of possible identity fraud in e-health studies and confirm the reliability of internet-based, SR anthropometric and demographic data collected in the Food4Me study. TRIAL REGISTRATION: NCT01530139 ( http://clinicaltrials.gov/show/NCT01530139 ).

Detection and classification of anomaly intrusion using hierarchy clustering and SVM

Anomaly detection as a kind of intrusion detection is good at detecting the unknown attacks or new attacks, and it has attracted much attention during recent years. In this paper, a new hierarchy anomaly intrusion detection model that combines the fuzzy c-means (FCM) based on genetic algorithm and SVM is proposed. During the process of detecting intrusion, the membership function and the fuzzy interval are applied to it, and the process is extended to soft classification from the previous hard classification. Then a fuzzy error correction sub interval is introduced, so when the detection result of a data instance belongs to this range, the data will be re-detected in order to improve the effectiveness of intrusion detection. Experimental results show that the proposed model can effectively detect the vast majority of network attack types, which provides a feasible solution for solving the problems of false alarm rate and detection rate in anomaly intrusion detection model.

A visual-numeric approach to clustering and anomaly detection for trajectory data

This paper proposes a novel application of Visual Assessment of Tendency (VAT)-based hierarchical clustering algorithms (VAT, iVAT, and clusiVAT) for trajectory analysis. We introduce a new clustering based anomaly detection framework named iVAT+ and clusiVAT+ and use it for trajectory anomaly detection. This approach is based on partitioning the VAT-generated Minimum Spanning Tree based on an efficient thresholding scheme. The trajectories are classified as normal or anomalous based on the number of paths in the clusters. On synthetic datasets with fixed and variable numbers of clusters and anomalies, we achieve 98 % classification accuracy. Our two-stage clusiVAT method is applied to 26,039 trajectories of vehicles and pedestrians from a parking lot scene from the real life MIT trajectories dataset. The first stage clusters the trajectories ignoring directionality. The second stage divides the clusters obtained from the first stage by considering trajectory direction. We show that our novel two-stage clusiVAT approach can produce natural and informative trajectory clusters on this real life dataset while finding representative anomalies.

Playing it safe : the influence of neighbourhood safety on children's physical activity − A review

Compared with previous generations, children spend less time playing outdoors and have lower participation rates in active transport. Many studies have identified lack of neighbourhood safety as a potential barrier to children's physical activity. This review describes concerns regarding ‘stranger danger’ and road safety, and discusses empirical studies that examine associations between neighbourhood safety and physical activity among youth. Variability of perceptions of safety between parents and youth are examined; ‘social traps’ are identified; and physical/social environmental interventions aimed at improving neighbourhood safety are discussed. A research agenda is suggested for further study of perceived and objective measures of neighbourhood safety and their associations with children's physical activity.

Reduction of adolescent alcohol use through family-school intervention: a randomized trial

Purpose
Two-year longitudinal follow-up data evaluated the behavioral impact of Resilient Families, a universal intervention that aimed to prevent early initiation and frequent and heavy adolescent alcohol use in secondary schools in Melbourne, Australia.

Methods
Of 24 secondary schools (62% of those approached), 12 were randomly assigned to intervention and 12 as controls. Intervention students received a social relationship curriculum; their parents received parent education handbooks and invitations to parent education events outlining strategies to encourage healthy adolescent development and reduce adolescent alcohol misuse. At Wave 1 (2004), students were in Year 7 secondary school (mean age, 12.3 years). Data were imputed for students completing at least two of three annual surveys (N = 2,354). Wave 3 (2006; mean, 14.5 years) main outcome measures for alcohol use were “any,” “frequent” (at least monthly), and “heavy” (five or more drinks in a session at least once in the prior fortnight). Multivariate logistic regression assessed intervention exposure effects, adjusting for school classroom clustering and baseline measures.

Results
Relative to controls, intervention students showed significant reductions in any lifetime use (adjusted odds ratio [AOR], .78; 95% confidence interval [CI], .62–.97), and reduced progression to frequent (AOR, .69; CI, .56–.86) and heavy use (AOR, .75; CI, .60–.94).

Conclusions
Randomized assignment to Resilient Families was associated with a significant reduction in adolescent alcohol use among families volunteering for the evaluation. Family–school-based interventions appear promising as a strategy to contribute to population reductions in currently high rates of adolescent alcohol misuse.

Perceived versus actual distance to local physical-activity facilities: does it really matter?

BACKGROUND: Research in adults shows poor agreement between self-reported and objectively measured proximity to physical activity resources; however there is little such research in adolescents. This study assessed the level of agreement between self-reported and objective measures of distance to physical activity resources in adolescents; and whether perceived or actual distance was related to actual use and physical activity levels. METHODS: 110 New Zealand high school students (12-18 years) were asked the time (in minutes) it would take them to walk from their home to the nearest physical activity resource, and whether they had used it in the previous month. The distance from participants' homes to the nearest resource was measured using GIS. Physical activity was assessed with accelerometers. RESULTS: Agreement was poor, with weighted Kappa Indices ranging from 0.1 to 0.4. The facilities used most frequently were schools (90%), public parks (76%), and playing fields (74%). Closer location was associated with higher use of some facilities only. Moderate-to-vigorous activity levels were not associated with self-reported or measured distance. CONCLUSIONS: Agreement between perceived and measured access is poor among adolescents. Further research is needed to understand how individual and social factors interact with environmental factors and whether improving awareness improves use.

Objective versus perceived walking distances to destinations : correspondence and predictive validity

Judgments concerning features of environments do not always correspond accurately with objective measures of those same features. Moreover, perceived and objectively assessed environmental attributes, including proximity of destinations, may influence walking behavior in different ways. This study compares perceived and objectively assessed distance to several different destinations and examines whether correspondence between objective and perceived distance is influenced by age, gender, neighborhood walkability, and walking behavior. Distances to most destinations close to home are overestimated, whereas distances to those farther away are underestimated. Perceived and objective distances to certain types of destinations are differentially associated with walking behavior. Perceived environmental attributes do not consistently reflect objectively assessed attributes, and both appear to have differential effects on physical activity behavior.

A systematic review of the validity and reliability of sedentary behaviour measures used with children and adolescents

The aim of this review was to evaluate the reliability and validity of methods used to assess the multiple components of sedentary behaviour (i.e. screen time, sitting, not moving and existing at low energy expenditure) in children and adolescents. Twenty-six studies met our inclusion criteria and were reviewed. Thirteen studies reported the reliability of self- and proxy-report measures of sedentary behaviour and seven of these were found to have acceptable test–retest reliability. Evidence for the criterion validity of self- and proxy-report measures was examined in three studies with mixed results. Seven studies examined the reliability and/or validity of direct observation and the findings were generally positive. Five studies demonstrated the utility of accelerometers to accurately classify sedentary behaviour. Self-report measures provide reliable estimates of screen time, yet their validity remains largely untested. While accelerometers can accurately classify participants’ behaviour as sedentary, they do not provide information about type of sedentary behaviour or context. Studies utilizing measures of sedentary behaviour need to more adequately report on the validity and reliability of the measures used.We recommend the use of objective measures of sedentary behaviour such as accelerometers, in conjunction with subjective measures (e.g. self-report), to assess type and context of behaviour.

Comparison of interpolation and central activation ratios as measures of muscle inactivation

The objective of this study was to investigate different methods of estimating muscle inactivation, derived from single and multiple voluntary contractions. Ten subjects performed maximal and submaximal leg extensor contractions to determine an interpolation (IT) or central activation ratio (CAR). A superimposed evoked force was compared with the force output of either a voluntary (CAR) or resting evoked contraction (IT ratio), or the ratios were inserted into regression equations (linear, polynomial, exponential). Linear-regression estimates of CAR using doublets and tetanus provided physiologically inaccurate values. Whereas IT ratios using doublets (IT-doublet) and tetanus (IT-tetanus) had a significant difference in only one interaction, IT-tetanus and CAR using a tetanus (CAR-tetanus) estimates provided the most extensive correlation within and between measures. Thus, tetanic stimulation superimposed upon single maximal or multiple contractions seems to provide the most valid measure of muscle inactivation when using the interpolated-twitch technique.

Web page clustering : a hyperlink-based similarity and matrix-based hierarchical algorithms

This paper proposes a hyperlink-based web page similarity measurement and two matrix-based hierarchical web page clustering algorithms. The web page similarity measurement incorporates hyperlink transitivity and page importance within the concerned web page space. One clustering algorithm takes cluster overlapping into account, another one does not. These algorithxms do not require predefined similarity thresholds for clustering, and are independent of the page order. The primary evaluations show the effectiveness of the proposed algorithms in clustering improvement.

Neighborhood walkability and the walking behavior of Australian adults

Background
The physical attributes of residential neighborhoods, particularly the connectedness of streets and the proximity of destinations, can influence walking behaviors. To provide the evidence for public health advocacy on activity-friendly environments, large-scale studies in different countries are needed. Associations of neighborhood physical environments with adults’ walking for transport and walking for recreation must be better understood.

Method
Walking for transport and walking for recreation were assessed with a validated survey among 2650 adults recruited from neighborhoods in an Australian city between July 2003 and June 2004, with neighborhoods selected to have either high or low walkability, based on objective measures of connectedness and proximity derived from geographic information systems (GIS) databases. The study design was stratified by area-level socioeconomic status, while analyses controlled for participant age, gender, individual-level socioeconomic status, and reasons for neighborhood self-selection.

Results
A strong independent positive association was found between weekly frequency of walking for transport and the objectively derived neighborhood walkability index. Preference for walkable neighborhoods moderated the relationship of walkability with weekly minutes, but not the frequency of walking for transport—walkability was related to higher frequency of transport walking, irrespective of neighborhood self-selection. There were no significant associations between environmental factors and walking for recreation.

Conclusions
Associations of neighborhood walkability attributes with walking for transport were confirmed in Australia. They accounted for a modest but statistically significant proportion of the total variation of the relevant walking behavior. The physical environment attributes that make up the walkability index are potentially important candidate factors for future environmental and policy initiatives designed to increase physical activity.

Associations of children's perceived neighborhood environments with walking and physical activity

Purpose. To examine associations between children's perceptions of the neighborhood environment and walking and physical activity.
Design. Cross-sectional study of a school-based sample.
Setting. Elementary schools in Melbourne, Australia.
Subjects. 280 children aged 10 years (response rate 78%).
Measures. A self-report survey assessed children's perceptions of the neighborhood physical and social environments and their weekly walking frequency. Physical activity was also objectively measured using accelerometers.
Results. Multiple linear regression analyses showed a positive association between walking frequency and the number of accessible destinations in the neighborhood among boys; having a neighborhood that was easy to walk/cycle around and perceiving lots of graffiti were positively associated with walking frequency among girls. Perceiving lots of litter and rubbish was positively associated with boys' overall physical activity, but no environmental variables were associated with girls' overall physical activity.
Conclusion. Several different environmental factors were associated with walking and physical activity. Perceptions of the neighborhood environment were more strongly associated with girls' walking than with objectively-measured physical activity. Future studies should confirm these findings using objective measures and prospective study designs.

Association of television viewing with fasting and 2-h postchallenge plasma glucose levels in adults without diagnosed diabetes

OBJECTIVE—We examined the associations of television viewing time with fasting plasma glucose (FPG) and 2-h postchallenge plasma glucose (2-h PG) levels in Australian adults.

RESEARCH DESIGN AND METHODS—A total of 8,357 adults aged >35 years who were free from diagnosed diabetes and who attended a population-based cross-sectional study (Australian Diabetes, Obesity and Lifestyle Study [AusDiab]) were evaluated. Measures of FPG and 2-h PG were obtained from an oral glucose tolerance test. Self-reported television viewing time (in the previous week) was assessed using an interviewer-administered questionnaire. Homeostasis model assessment (HOMA) of insulin sensitivity (HOMA-%S) and ß-cell function (HOMA-%B) were calculated based on fasting glucose and insulin concentrations.

RESULTS—After adjustment for confounders and physical activity time, time spent watching television in women was positively associated with 2-h PG, log fasting insulin, and log HOMA-%B and inversely associated with log HOMA-%S (P < 0.05) but not with FPG. No significant associations were observed with glycemic measures in men. The ß-coefficients across categories of average hours spent watching television per day (<1.0, 1.0–1.9, 2.0–2.9, 3.0–3.9, and ≥4.0) for 2-h PG in women were 0 (reference), 0.009, 0.047, 0.473, and 0.501, respectively (P for trend = 0.02).

CONCLUSIONS—Our findings highlight the unique deleterious relationship of sedentary behavior (indicated by television viewing time) and glycemic measures independent of physical activity time and adiposity status. These relationships differed according to sex and type of glucose measurement, with the 2-h PG measure being more strongly associated with television viewing. The findings suggest an important role for reducing sedentary behavior in the prevention of type 2 diabetes and cardiovascular disease, especially in women.