Abstract: Research in predictive maintenance modeling has improved in the recent years to predict failures and needed maintenance with high accuracy, saving cost and improving manufacturing efficiency. However, classic prediction models provide little valuable insight towards the most important features contributing to the failure. By analyzing and quantifying feature importance in predictive maintenance models, cost saving can be optimized based on business goals. First, multiple classifiers are evaluated with cross-validation to predict the multi-class of failures. Second, predictive performance with features provided by different feature selection algorithms are further analyzed. Third, features selected by different algorithms are ranked and combined based on their predictive power. Finally, linear explainer SHAP (SHapley Additive exPlanations) is applied to interpret classifier behavior and provide further insight towards the specific roles of features in both local predictions and global model behavior. The results of the experiments suggest that certain features play dominant roles in predictive models while others have significantly less impact on the overall performance. Moreover, for multi-class prediction of machine failures, the most important features vary with type of machine failures. The results may lead to improved productivity and cost saving by prioritizing sensor deployment, data collection, and data processing of more important features over less importance features.
Abstract: An autonomous environmental monitoring system
(Smart Landfill) has been constructed for the quantitative
measurement of the components of landfill gas found at borehole
wells at the perimeter of landfill sites. The main components of
landfill gas are the greenhouse gases, methane and carbon dioxide
and have been monitored in the range 0-5 % volume. This monitoring
system has not only been tested in the laboratory but has been
deployed in multiple field trials and the data collected successfully
compared with on-site monitors. This success shows the potential of
this system for application in environments where reliable gas
monitoring is crucial.
Abstract: In order to monitor for traffic traversal, sensors can be
deployed to perform collaborative target detection. Such a sensor
network achieves a certain level of detection performance with the
associated costs of deployment and routing protocol. This paper
addresses these two points of sensor deployment and routing algorithm
in the situation where the absolute quantity of sensors or total energy
becomes insufficient. This discussion on the best deployment system
concluded that two kinds of deployments; Normal and Power law
distributions, show 6 and 3 times longer than Random distribution in
the duration of coverage, respectively. The other discussion on routing
algorithm to achieve good performance in each deployment system
was also addressed. This discussion concluded that, in place of the
traditional algorithm, a new algorithm can extend the time of coverage
duration by 4 times in a Normal distribution, and in the circumstance
where every deployed sensor operates as a binary model.