Browsing by Author "Tiwari, Ashutosh"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
- Circular business models in high value manufacturing: Five industry cases to bridge theory and practiceOkorie, Okechukwu; Charnley, Fiona; Russell, Jennifer D.; Tiwari, Ashutosh; Moreno, Mariale (2021-01-12)The transition to a circular economy (CE) requires companies to evaluate their resource flows, supply chains, and business models and to question the ways in which value is created. In the high value manufacturing (HVM) sector, this evaluation is critical, as HVM enables value in nonconventional forms, beyond profit, including unique production processes, brand recognition, rapid delivery times, and highly customized services. We investigate the role of value, cost, and other factors of influence in the selection of a circular business model (CBM) for HVM. Explored through five case studies using a qualitative evaluation of circularity, we then contribute to the emerging field of CBMs by modifying the CBM canvas that can capture the nontraditional value, traditional value, cost, and other influencing factors enabled via CBM adoption in HVM. Finally, the important role of digital technologies for incentivizing and enabling CBM adoption, is clarified.
- A novel machine learning and deep learning semi-supervised approach for automatic detection of InSAR-based deformation hotspotsTiwari, Ashutosh; Shirzaei, Manoochehr (Elsevier, 2024-02)Over the past two decades, Interferometric synthetic aperture radar (InSAR) has been invaluable for studying earth surface deformation and related effects. Deformation maps generated through multi-temporal InSAR processing methods are however difficult to interpret accurately by general individual users, decision-makers, and non-domain experts owing to the volume, variety, and velocity they are produced. This paper proposes a semi-supervised machine learning based information mining approach to simplify these deformation maps and detect hotspots by extracting prominent signals from time series deformation. The approach initially combines two machine learning based clustering methods named time series k-means (TSKM) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithms to derive clusters with unique spatiotemporal deformation behavior, using time series deformation output generated from Wavelet-based InSAR (WabInSAR) method. Clustering results generated from this unsupervised machine learning approach are later used as training labels to develop two deep learning models, one using long short term memory (LSTM) networks alone and another using a combination of LSTM and single-layer perceptron for supervised training. The developed LSTM and LSTM + Perceptron models efficiently learn from the cluster labels, reaching an accuracy of 97.3 %. Further, the deep learning models significantly reduce the computational time from orders of days (∼5) to hours (∼2) while training and from hours to minutes during prediction. We evaluate the developed approach over Los Angeles, a highly challenging area affected by umpteen deformation events that are challenging to categorize. The outcome of the proposed approach produces hotspots of deforming areas in Los Angeles, providing a generalized and more precise picture of events, much appreciable to non-domain experts. The approach can augment any of the multi-temporal InSAR processing chains and is applicable to different deformation prone sites, aiding in derivation of deformation hotspots from time series deformation maps.
- SharkPulse Validator GameKothari, Aman; Patel, Feneel; Raya, Ray; Shroff, Tirth; Tiwari, Ashutosh (Virginia Tech, 2021-12-15)SharkPulse is an initiative to involve citizen scientists in monitoring global shark populations. It is inspired by Stanford’s Shark Baseline Project, and it aims to collect image-based sightings of sharks from around the world to support research on ecology and conservation and increase public awareness of their conservation status. This project is currently headed by Dr. Francesco Ferretti. He is an Assistant Professor for the Department of Fish and Wildlife Conservation. The team was provided with a platform to update and implement new design changes to the original website. The original website was built using WordPress for frontend CSS and HTML, with RShiny providing the backend to implement the Validation Monitor. We were tasked with converting the static framework of the website to a more dynamic and responsive framework, improving on the current gamification scheme with a better rewards system and incentives, sourcing data from streamlined pipes, and developing a convenient user authentication system across multiple social media platforms to allow users to log in and store their points. This report contains information on the original project given to the team, and the changes the team implemented as per Dr. Ferretti's request. The report also highlights work that can be attempted by future teams that build on the current project as worked on and delivered by the team.