Tentative title of my PhD thesis: "Big data–driven machine-learning approach for improvement in the field of health logistics"
My main supervisor is Associate Professor Berit Irene Helgheim and the co-supervisors are Postdoctoral Fellow Birgithe Eckermann Sandbæk and Assistant Professor Nikhil Varma.
Background
I am currently a PhD student in logistics at Molde University College, Specialized University in Logistics (MUC).I have Bachelor of Technology degree in Mechanical Engineering from Amrita Vishwa Vidyapeetham University and Master of Science degree in Engineering Logistics from Molde University College.
Teaching work:
-
IBE700 Enterprise Resource Planning Systems (ERP-systems) - Spring 2019
-
LOG904-117 Business Intelligence with SAP - Fall 2018, Fall 2019
Further details:
All relevant info on my LinkedIn page
Emneord:
Stipendiat Logistikk
Publikasjoner
-
Aravazhi, Agaraoli
(2021).
Hybrid machine learning models for forecasting surgical case volumes at a hospital.
AI.
2(4),
s. 512–526.
doi:
10.3390/ai2040032.
Vis sammendrag
Abstract
Recent developments in machine learning and deep learning have led to the use of multiple algorithms to make better predictions. Surgical units in hospitals allocate their resources for day surgeries based on the number of elective patients, which is mostly disrupted by emergency surgeries. Sixteen different models were constructed for this comparative study, including four simple and twelve hybrid models for predicting the demand for endocrinology, gastroenterology, vascular, urology, and pediatric surgical units. The four simple models used were seasonal autoregressive integrated moving average (SARIMA), support vector regression (SVR), multilayer perceptron (MLP), and long short-term memory (LSTM). The twelve hybrid models used were a combination of any two of the above-mentioned simple models, namely, SARIMA–SVR, SVR–SARIMA, SARIMA–MLP, MLP–SARIMA, SARIMA–LSTM, LSTM–SARIMA, SVR–MLP, MLP–SVR, SVR–LSTM, LSTM–SVR, MLP–LSTM, and LSTM–MLP. Data from the period 2012–2018 were used to build and test the models for each surgical unit. The results indicated that, in some cases, the simple LSTM model outperformed the others while, in other cases, there was a need for hybrid models. This shows that surgical units are unique in nature and need separate models for predicting their corresponding surgical volumes. View Full-Text
Keywords: time series, seasonal autoregressive integrated moving average, machine learning, hybrid model, demand, hospital, surgical unit
-
Aravazhi, Agaraoli; Helgheim, Berit & Jæger, Bjørn
(2020).
Investigation of the influence of product variety on inventories in hospitals.
Engineering Management in Production and Services.
ISSN 2543-6597.
12(1),
s. 34–44.
doi:
10.2478/emj-2020-0003.
Fulltekst i vitenarkiv
Vis sammendrag
The literature on product variety provides great insights into various businesses.
However, little research has been conducted on product variety in the healthcare
industry. This study aims to explore the influence of product variety on inventory
in hospitals. Since most hospitals are known to replenish products using a homegrown
ad-hoc system, a model is developed for exploring all possible product combinations
and substitutions. This article presents the behaviour of product substitution, which
may be either one-to-one or many-to-one for both sterile and non-sterile products,
in the hospital with cost factors. It discusses the product variety reduction and its
corresponding cost impacts. The data on a hospital inventory over the course of six
years has been procured from a hospital in Norway. Based on the results, the hospital
could have a potential product variety reduction of approximately 11% and cost savings
from the spending of approximately NOK 3.6 million. Reducing the variety of products
in hospital inventories proves to be an approach to reducing costs. The model
developed for the research is universal in nature and could be used in other fields, such
as retail, marketing etc.
-
Müller, Falko & Aravazhi, Agaraoli
(2020).
A new generalized travel cost based connectivity metric applied to Scandinavian airports.
Transportation Research Part D: Transport and Environment.
ISSN 1361-9209.
81(April ),
s. 1–21.
doi:
10.1016/j.trd.2020.102280.
Vis sammendrag
This article proposes a new, generalized travel cost based method to operationalize network accessibility provided by airports. The approach is novel as it integrates features of network topology with multiple quality aspects of scheduled air transport services into one metric. The method estimates generalized travel costs for the full set of feasible travel paths between an airport and all network destinations. Rooftop modeling accounts for schedule delay and isolates the most cost-efficient travel paths per O-D relation. Respecting the assumed arrival time preference of passengers and adjusting for destination importance, connectivity scores are derived. The method is then applied to explore changes in the global connectivity pattern of Scandinavian airports from 2004 to 2018. The results suggest distinct spatial differences throughout the network, but less pronounced in size than suggested by popularly applied connectivity measures. Findings also highlight the importance of the geographical location as a determinate of an airport’s connectivity.
Keywords: Air transport; Accessibility; Connectivity; Network topology; Air service features.
Se alle arbeider i Cristin
Publisert 3. sep. 2018 15:09
- Sist endret 10. jan. 2020 15:19