Downloads: 14
Research Paper | Information Technology | Volume 8 Issue 10, October 2019 | Pages: 1860 - 1864 | United States
Optimizing Fleet Performance: A Deep Learning Approach on AWS IoT and Kafka Streams for Predictive Maintenance of Heavy - Duty Engines
Abstract: Predictive maintenance (PdM) predicts machine failures in heavy - duty vehicles with diesel engines. PdM utilizes deep learning algorithms on vast amounts of Internet of Things (IoT) data to forecast potential failures accurately. However, the sheer magnitude and rapidity of data generated makes this process incredibly expensive. We propose a novel model executed on Amazon Web Services (AWS) IoT and Kafka Streams to mitigate this challenge. Through our extensive experiments, we confidently demonstrate the effectiveness and efficiency of our approach, including the successful implementation of the activation threshold parameter, resulting in significantly enhanced prediction accuracy. Moreover, we introduce a valuable assessment (VA) method for evaluating the incidence rate scale, further enhancing our predictive capabilities. The results obtained from our comprehensive analysis highlight the superior performance achieved through a meticulously balanced VATP and VA strategy, establishing our solution as a game - changer in predictive maintenance for heavy - duty vehicles.
Keywords: Predictive Maintenance, Heavy - Duty Engine, Industry 4.0, Internet of Things (IoT), Artificial Intelligence (AI), Machine Learning (ML), Smart Manufacturing (SM)
How to Cite?: Vishwanadham Mandala, "Optimizing Fleet Performance: A Deep Learning Approach on AWS IoT and Kafka Streams for Predictive Maintenance of Heavy - Duty Engines", Volume 8 Issue 10, October 2019, International Journal of Science and Research (IJSR), Pages: 1860-1864, https://www.ijsr.net/getabstract.php?paperid=ES24516094655, DOI: https://dx.doi.org/10.21275/ES24516094655