Posts by Collection

portfolio

Soil Classification using Machine Learning: From First Principles

Published:

This project is on understanding and developing intution for Machine Learning and Neural Networks. These techniques/approaches cannot be and should not be treated as ‘Black-Boxes’. Inthis project, I have tried to show this fact visually by selecting a not much explored dataset of Soil Classification.

publications

Least Cost Path Pipeline Routing Using Spatial Multi-criteria Analysis for Vidarbha Region: A Case Study

Published in Book Chapter in Geospatial and Soft Computing Techniques, 2023

The Western Vidarbha region of Maharashtra state is experiencing drought, whereas the Eastern Vidarbha region is experiencing floods. National Water Development Agency (NWDA) proposed an intra-state link project to divert surplus water from the Wainganga basin (Eastern Vidarbha) to the Nalganga reservoir (Western Vidarbha) via canals. This study investigates possible alternative routes for intra-state link via pipeline, using spatial multi-criteria analysis (SMCA) and least cost path (LCP) approach in the GIS environment. In this study, SMCA implementing analytical hierarchal process (AHP), was used to combine different spatial data coherently. This approach also incorporates the subjective opinion of the decision-maker into the route planning process. Pathfinding was carried out using the LCP approach implementing Dijkstra’s Algorithm. The route planning process considered four main factors viz. economy, engineering, environmental, and social. These factors in turn comprised of different criteria and sub-criteria such as slope, soil type, geomorphology, land use/land cover, drainage network, and settlements, which were structured in a hierarchal fashion. Four alternative routes were generated which prioritized one of the four main factors. These routes were then evaluated and ranked using the AHP method, and the optimal route is suggested. It was found that the total length and number of road-rail-river intersections for LCP generated route were, respectively, 10.7% and 5.88% less than NWDA proposed route. The average ground slope for the LCP-generated route was 6.22% and 12.15% for NWDA-proposed route. The results show that incorporating SMCA and LCP for route planning in a GIS environment generates results better than the conventional approach. Further studies are needed to investigate the utility of anisotropic cost surfaces or the use of alternative LCP algorithms such as the A* algorithm or the development of local route optimization techniques.

Recommended citation: Mhamane, A., Mirajkar, A.B. (2023). Least Cost Path Pipeline Routing Using Spatial Multi-criteria Analysis for Vidarbha Region: A Case Study. In: Timbadiya, P.V., Patel, P.L., Singh, V.P., Mirajkar, A.B. (eds) Geospatial and Soft Computing Techniques. HYDRO 2021. Lecture Notes in Civil Engineering, vol 339.Springer, Singapore https://link.springer.com/chapter/10.1007/978-981-99-1901-7_30

talks

teaching