Avin D. Mathew

Empirical analysis of support vector machine ensemble classifiers

Ensemble classification – combining the results of a set of base learners – has received much attention in the machine learning community and has demonstrated promising capabilities in improving classification accuracy. Compared with neural network or decision tree ensembles, there is no comprehensive empirical research in support vector machine (SVM) ensembles. To fill this void, this paper analyses and compares SVM ensembles with four different ensemble constructing techniques, namely bagging, AdaBoost, Arc-X4 and a modified AdaBoost. Twenty real-world data sets from the UCI repository are used as benchmarks to evaluate and compare the performance of these SVM ensemble classifiers by their classification accuracy. Different kernel functions and different numbers of base SVM learners are tested in the ensembles. The experimental results show that although SVM ensembles are not always better than a single SVM, the SVM bagged ensemble performs as well or better than other methods with a relatively higher generality, particularly SVMs with a polynomial kernel function. Finally, an industrial case study of gear defect detection is conducted to validate the empirical analysis results.

A Review of the MIMOSA OSA-EAI Database for Condition Monitoring Systems

Condition monitoring systems are prevalent in industries that contain critical assets. Their use allows the detection of potential failures at an early stage in order to minimise downtime and maintenance costs of a system. Many condition monitoring systems are commercially available, however these systems are limited in their support of diagnostic and prognostic models developed by external parties. An attempt to develop an extensible condition monitoring software lead to the creation of an ISO 13374/OSA-CBM based system named BUDS, which focuses on vibration condition monitoring. Research into condition monitoring data management strategies led our team to implement a database to support the data-driven application. A database based on the MIMOSA OSA-EAI database specification was used to store asset and sensor location, measurement event, raw sensor and processed signal data. This paper investigates the use of the MIMOSA OSA-EAI database for condition monitoring systems, and presents several issues encountered during the development process.

Renewal decision support for linear assets

Linear (or continuous) assets are engineering infrastructure that usually spans long distances and can be divided into different segments, all of which perform the same function but may be subject to different loads and environmental conditions. Typical linear assets include railway lines, roads, pipelines and cables. How and when to renew such assets are critical decisions for asset owners as they normally involves significant capital investment. Through investigating the characteristics of linear asset renewal decisions and identifying the critical requirements that are associated with renewal decisions, we present a multi-criteria decision support method to help optimise renewal decisions. A case study that concerns renewal of an economiser’s tubing system in a coal-fired power station is adopted to demonstrate the application of the method. Although the paper concerns a particular linear asset decision type, the approach has broad applicability for linear asset management generally.

Bringing the MIMOSA OSA-EAI into an Object-Oriented World

Despite being poised as a standard for data exchange for operations and maintenance data, the database heritage of the MIMOSA OSA-EAI is clearly evident from using a relational model at its core. The XML schema (XSD) definitions, which are used for communication between asset management systems, are based on the MIMOSA common relational information schema (CRIS), a relational model, and consequently, many database concepts permeate the communications layer. The adoption of a relational model leads to several deficiencies, and overlooks advances in object-oriented modelling and XSD itself. A decision was made to adopt an object-oriented approach for an upcoming version of the specification, and the common conceptual object model (CCOM) sees a transition to fully utilising object-oriented features for the standard. Unified modelling language (UML) is used as a medium for documentation as well as facilitating XSD code generation. This paper details some of the decisions faced in developing the CCOM and provides a glimpse into the future of asset management data exchange standards.

Open standards-based system integration for asset management decision support

Over the last decade, system integration has grown in popularity as it allows organisations to streamline business processes. Traditionally, system integration has been conducted through point-to-point solutions – as a new integration scenario requirement arises, a custom solution is built between the relevant systems. Bus-based solutions are now preferred, whereby all systems communicate via an intermediary system such as an enterprise service bus, using a common data exchange model. This research investigates the use of a common data exchange model based on open standards, specifically MIMOSA OSA-EAI, for asset management system integration. A case study is conducted that involves the integration of processes between a SCADA, maintenance decision support and work management system. A diverse number of software platforms are employed in developing the final solution, all tied together through MIMOSA OSA-EAI-based XML web services. The lessons learned from the exercise are presented throughout the paper.

Implementation of Asset Health Assessment System with Pattern-Oriented Design and Practice

To integrate condition monitoring and reliability analysis with asset life-cycle management, a general-purpose asset health assessment system architecture is being developed. This system will assist management and maintenance staff to make decisions based on condition and reliability data, scientific analysis, and integrated information thereby reducing the asset management cost for enterprises. In this paper, a pattern-oriented software architecture design is presented. The functionality of the system is examined and some issues are identified. In particular, the paper elaborates on how design patterns are applied in the implementation of the different tiers in the system. It also presents a discussion detailing the system architecture that facilitates extensibility, flexibility, and robustness of the software system. To simplify the implementation of design patterns, various methodologies based on the in-built features of Microsoft .NET framework are applied

Creating an asset registry for railway electrical traction equipment with open standards

An asset registry arguably forms the core system that needs to be in place before other systems can operate or interoperate. Most systems have rudimentary asset registry functionality that store assets, relationships, or characteristics, and this leads to different asset management systems storing similar sets of data in multiple locations in an organisation. As organisations have been slowly moving their information architecture toward a service-oriented architecture, they have also been consolidating their multiple data stores, to form a “single point of truth”. As part of a strategy to integrate several asset management systems in an Australian railway organisation, a case study for developing a consolidated asset registry was conducted. A decision was made to use the MIMOSA OSA-EAI CRIS data model as well as the OSA-EAI Reference Data in building the platform due to the standard’s relative maturity and completeness. A pilot study of electrical traction equipment was selected, and the data sources feeding into the asset registry were primarily diagrammatic based. This paper presents the pitfalls encountered, approaches taken, and lessons learned during the development of the asset registry.