Search

An essential requirement in any data analysis is to have a response variable representing the aim of the analysis. Much academic work is based on laboratory or simulated data, where the experiment is controlled, and the ground truth clearly defined. This is seldom the reality for equipment performance in an industrial environment and it is common to find issues with the response variable in industry situations. We discuss this matter using a case study where the problem is to detect an asset event (failure) using data available but for which no ground truth is available from historical records. Our data frame contains measurements of 14 sensors recorded every minute from a process control system and 4 current motors on the asset of interest over a three year period. In this situation the ``how to'' label the event of interest is of fundamental importance. Different labelling strategies will generate different models with direct impact on the in-service fault detection efficacy of the resulting model. We discuss a data-driven approach to label a binary response variable (fault/anomaly detection) and compare it to a rule-based approach. Labelling of the time series was performed using dynamic time warping followed by agglomerative hierarchical clustering to group events with similar event dynamics. Both data sets have significant imbalance with 1,200,000 non-event data but only 150 events in the rule-based data set and 64 events in the data-driven data set. We study the performance of the models based on these two different labelling strategies, treating each data set independently. We describe decisions made in window-size selection, managing imbalance, hyper-parameter tuning, training and test selection, and use two models, logistic regression and random forest for event detection. We estimate useful models for both data sets. By useful, we understand that we could detect events for the first four months in the test set. However as the months progressed the performa

© 2017 Elsevier Ltd Adherence to procedures is critical to the safety and performance of maintenance tasks; however, few studies of procedure compliance among maintenance personnel have been reported. The present study evaluated a theoretical model in which management approaches to procedure compliance were linked to compliance outcomes through user perceptions of positive and negative procedure attributes. New scales were developed to assess these variables; hypotheses derived from the model were tested in survey data collected from maintainers in the mining industry (N = 176). A structural equation model showed acceptable fit statistics; findings were broadly consistent with the initial hypotheses. As predicted, positive and negative dimensions of procedure attributes and compliance/non-compliance were perceived as distinct constructs, and were implicated in different pathways of the model. Also supporting the initial hypotheses, user involvement and managers’ learning-oriented responses to non-compliance were linked to favourable compliance outcomes through perceived procedure attributes. Learning-oriented responses were also directly associated with greater compliance. In addition, and contrary to prediction, punitive management responses positively predicted compliance. As discussed in the paper, these findings contribute new insights, relevant in both research and industry contexts, to understanding procedure compliance among maintainers.

© 2018 Elsevier Ltd This research presents a model of employee behaviour related to the use of procedures in safety critical industries. A key contribution is the focus on procedure-related behaviour that is enacted when employees are engaged with their work—such as when they invest personal effort into complying with procedures and voice suggestions or concerns they have with the procedures. This study examines how these two engaged behaviours, namely, effort towards compliance and procedure-related voice behaviour, are influenced by psychological and organizational factors. In doing so, we introduce Kahn's theory of work engagement into the safety procedure literature. Survey data were collected from 152 maintainers in a mining corporation in Australia. The data were analysed using path analysis. Our results indicated that supervisor helping behaviour had a significant positive effect on effort towards compliance and procedure-related voice, via its influence on perceived usefulness and job self-efficacy, respectively. The results suggest that employees’ perceptions of the utility value of the procedures as well as their own capabilities in carrying out their job tasks play a key role in shaping how employees use procedures. Organizations should influence those perceptions in order to encourage employees’ engagement in the use of procedures. Our study suggests that this can be achieved by structuring the role of supervisors to ensure they have the capacity and availability to help their team members.

© Copyright 2017, AIDIC Servizi S.r.l. Aim: The paper outlines a systemic approach to understanding and assessing safety capability in high-risk industries, like off-shore oil, gas industry, chemical operators. The "Fitness to Operate" framework (acronym: FTO) (Griffin et al., 2014) has been recently defined by three enabling capitals that create safety capability: organizational capital, social capital, and human capital. Furthermore, each type of capital is identified by more specific dimensions based on current theories of safety, management, and organizational processes. In this paper, we will present a multidimensional assessment tool that offers a comprehensive picture of safety capability by real industrial operators in order to understand and evaluate their "fitness-to-operate" (FTO). Method: This current paper aims to describe the multi-phase development process of a FTO assessment tool in the format of a multidimensional survey questionnaire. A) The first research phase consists of the item generation of a large prototype pool with about 200 contents-items covering the 27 dimensions of the conceptual representation of the FTO framework. This initial pool was developed by a team of academic researchers, through a deductive process, and in the light of the original FTO conceptualization, as defined by Griffin and colleagues (2014) B) In a second research phase, the initial pools of items were re-examined by a new pool of academic researchers, assessing the quality of the contents, and in order to refine the extensive version of the prototype, eliminating potential redundancies and inadequate items. C) In a third phase with structured interviews to a pool of industrial experts (senior safety managers; senior executives), the authors assessed the quality of the prototype tool developed by the academic researchers, in order to evaluate and ranks the items of the prototype in term of quality, in order to define and identify a shorter version of the prototype. Al

Setting targets is a common way for organisations to establish performance expectations. However the validity of targets is challenged when performance is influenced by factors beyond the control of the manager. This project examines the issue of target setting for a single asset performance measure across different geographical areas. The performance measure is “Wastewater Blockages per 100km” at the Water Corporation (WC). Factors relating to these blockages are examined for eight operating locations with quite different characteristics. A generalised linear regression model is developed to determine the influence of age, pipe type, network length, population, rainfall as well as costs associated with preventative and corrective maintenance. Preliminary results reinforce that age and pipe type significantly influence blockage rates and this confirmation can be used to develop an understanding of how different districts might perform. A key result is that there is no relationship inside a district between management-controlled activities such as preventative maintenance and blockages. Unless a relationship can be established it would be premature to set targets for managers as the variables they can control (costs) do not appear to make an impact on what they are being assessed against., Structural Engineering, Civil Engineering and Geosciences

Cavitation is one of the main problems reducing the longevity of centrifugal pumps in industry today. If the pump operation is unable to maintain operating conditions around the best efficiency point, it can be subject to conditions that may lead to vaporisation or flashing in the pipes upstream of the pump. The implosion of these vapour bubbles in the impeller or volute causes damaging effects to the pump. A new method of vibration cavitation detection is proposed in this paper, based on adaptive octave band analysis, principal component analysis and statistical metrics. Full scale industrial pump efficiency testing data was used to determine the initial cavitation parameters for the analysis. The method was then tested using vibration measured from a number of industry pumps used in the water industry. Results were compared to knowledge known about the state of the pump, and the classification of the pump according to ISO 10816.

Cavitation is a major problem facing centrifugal pumps in industry today. Unable to constantly maintain operating conditions around the best efficiency point, centrifugal pumps are subject to conditions that may lead to vaporisation or flashing in the pipes upstream of the pump. The implosion of these vapour bubbles in the impeller or volute causes damaging effects to the pump. A new method of cavitation detection is proposed in this paper based on spectral methods. Data used to determine parameters were obtained under ideal conditions, while the method was tested using industry acquired data. Results were compared to knowledge known about the state of the pump, and the classification of the pump according to ISO 10816.

Cavitation is a major problem facing centrifugal pumps in industry today. Unable to constantly maintain operating conditions around the best efficiency point, centrifugal pumps are subject to conditions that may lead to vaporisation or flashing in the pipes upstream of the pump. The implosion of these vapour bubbles in the impeller or volute causes damaging effects to the pump. A new method of cavitation detection is proposed in this paper based on spectral methods. Data used to determine parameters were obtained under ideal conditions, while the method was tested using industry acquired data. Results were compared to knowledge known about the state of the pump, and the classification of the pump according to ISO 10816.

The international asset management standard ISO 55001, introduced in early 2014, outlines the requirement for an effective Asset Management System. Asset Management practitioners are seeking guidance on implementing one of the key requirements of the standard: the “line of sight” between the Corporate, Asset Management objectives and its relevant performance measures. This alignment ensures regulatory compliance, improved communication, informed asset investment decisions, managed risks and increased operational effectiveness. This paper demonstrates that a ‘line of sight’ is achievable through the application of the Balanced Scorecard approach using the Asset Management function at the Water Corporation as an example. The approach is deployed across two phases: the development of Asset Management Objectives through a consultative Asset Strategy Mapping exercise; and the selection of a balanced set of performance measures that link to the Strategy Map. The result of this approach is the creation of the ‘Asset Planning Performance Measurement Framework’. This framework is tested using water utility data resulting in the realisation of a ‘line of sight’ between asset performance measures and corporate objectives., Structural Engineering, Civil Engineering and Geosciences

Cavitation is one of the major problems associated with the operation of centrifugal pumps. Cavitation occurs when vapour bubbles that are formed due to a drop in pressure in the pipes upstream of the centrifugal pump implode under the added pressure within the volute of the pump. These implosions wear away the impeller, and sometimes the volute itself, which if left unchecked, would render the pump inoperable. Much research has been done in the detection of cavitation through: indicators in certain audible frequencies, drop in the net positive suction head, visual inspection using a transparent casing and a stroboscopic light, paint erosion inside the volute and on the impeller, changes in pressure within the flow or volute, and vibration within certain frequency ranges. Vibration detection is deemed as one of the more difficult methods due to other structural and environmental factors that may influence which frequencies may be present during the onset of cavitation. Vibration measurement however is most easily measured and deployable in an automated condition monitoring scenario.It is proposed that an increasing trend in a set of statistical parameters, rather than a firm threshold of a single parameter, would provide a robust indication for the onset of cavitation. Trends in these statistical parameters were obtained from data collected on a pump forced to cavitate under several different operating conditions. A single cavitation indicator is outlined utilizing these statistical parameters that can quantify the level of cavitation in a centrifugal pump.