OBJECTIVE

EDUCATION

• Master of Science: Operations Research, IITM
• Bachelor of Engineering: Computer Science and Engineering

SKILLS

• Advanced skills in Microsoft Office including Excel and Access.
• Statistics/Simulation software: ARENA, SPSS and SAS.
• Computer software: C, C++, VC++, Java, HTML, Oracle and SQL.

PROJECTS

• Take Solutions - Sales Forecasting (Individual Project - 10 months)
• Identified the issues with the inbuilt forecasting technique of the client.
• Realized that there are a number of external/environmental factors affecting the sales.
• Structured and implemented Neural Network technique and ended up with 30% forecasting accuracy.
• Identified the issues with Neural Network for this specific customer product forecast.
• Proposed Exponential Smoothing as a better forecasting technique for the customer.
• When coded and implemented the Exponential Smoothing technique, it was found that the forecasting accuracy of the products improved to an overall of 80% (in all of the 183 SKUs in each of the 4 regions of the country).
• Visteon, Chennai, India - Delay in delivery of the Finished Product (Team Work - 3 months)
• Understood the material flow and information flow of the Organization.
• Analyzed the issues that relate to delay in delivery of the product.
• Studied about each of the individual issue in detail to identify the root cause.
• Identified the need for a separate Production Planning Department and Forecasting Team.
• Calculated the optimal order quantity by forecasting the annual demand.
• Provided insights about the analysis to the Top Management and concluded with solutions that can be taken in order to improve the situation in the organization.
• When implemented a few of the suggested solutions, the delay in product delivery decreased by 40%.
• Simpson & Co. Limited, Chennai - Re-Design the process workflow (Individual Work - 3 months)
• Studied the manufacturing process of the organization and the detailed function of each of the 5 Departments - Stores, Quality, Planning, Production and Finance department (Logistics was outsourced)
• Identified the Quality department inspects the raw material one by one which in most cases is below the acceptance level and hence turns out to be a time consuming process.
• Re-Designed an effective alternative workflow for the organization
• Suggested that the raw materials could be inspected by random sampling method (as the suppliers test for the quality of the materials). In case of any quality issues above the acceptable range were identified, and then a penalty cost could be imposed so that the supplier will be extra conscious with their quality.
• MindTree, Bangalore, India (Team Project - 4 months)
• Studied the working of the other softwares like Bit Torrent.
• Satisfied the requirements of the organization by developing a Client side Torrent downloader using J2EE for the internal working of the organization.
• Experienced a real IT industry scenario of working with documentations, weekly-monthly status reports and so on.

RESEARCH WORK
Thesis: Modifications in exponentially weighted and neural network based methods for time series data including explanatory variables.
Keywords: Forecasting, Time Series Data, Exponential Smoothing, Neural Networks, Casual Model, Explanatory Variables/Exogenous Variables.
Abstract: The goal of Time Series Forecasting (TSF) is to predict the system's behavior from only past data. Although the traditional TSF methods give better forecasts, they become handicap with noisy or nonlinear components. The assumption is that a model may not be sufficient to represent the complete behavior of time series, specifically in case of both linear and nonlinear patterns. Hence the hybrid model is to be developed and tested for a varied set of erratic inputs, while observing their effects on the sales variance.

Seasonality Adjusted Exponential Smoothing (SAES):

• Variant of the Holt-Winters’ model with difference in the calculation of seasonality index is developed.
• Solutions prove that the SAES is better than the linear models (HWA and HWM).
• SAES can perform equally as good as nonlinear model (Artificial Neural Network - ANN model).

Seasonality Adjusted Neural Network (SANN):

• A model is developed by incorporating nonlinearity through a SANN model.
• The linear part of the SANN model is used to forecast the sales.
• The nonlinear model of the SANN model is used to forecast the residual part of the linear portion.
• Model is tested with real life data set.
• Results are compared with a linear model (SAES) and nonlinear models including an ANN model and a double artificial neural network (DANN) model.
• Solutions provided shows that the SANN model performs better compared to the other models.

ANN with Explanatory Variables:

• A nonlinear neural network model is developed as a causal and time series model to further improve the forecasting accuracy.
• Model is validated with data sets and the results are provided for the specific data sets considered.
• The forecasting accuracy of the model considering the explanatory variable is found to be better in terms of forecasting error (minimized forecast error) in many cases.

Overall Result: The results for all the models are tabulated and analyzed. It is concluded that SANN model perform well (compared to the other 2 models developed) in real life situations.
M.S. COURSE WORK

• Operations Management
• Operations Research
• Services and Operations Management
• Logistics and Distribution Management
• Supply Chain Management
• Manufacturing Systems Management
• Data Analysis for Management
• Strategic Sourcing
• Simulation Modeling and Applications
• Systems Thinking
• Probability and Statistics
• Business Research Methodology

TRAINING/WORKSHOPS ATTENDED

• Short-Term Course on "Multi Criteria Decision Making" conducted by Prof. A. Ravi Ravindran, Pennsylvania State University.
• Statistics workshop titled "What Statistical test and why?" conducted by Mr. Ram Vishwanathan, Professor and a Research consultant for IBM.
• Workshop on "ARENA" conducted by Prof. Shahabudeen, Head of Department of Industrial Engineering, Anna University.
• Workshop on "Multivariate Statistics" conducted by Prof. Punniyamoorthy, NIT Trichy.
• Seminar by Dr. R.Chidambaram Principal Scientific Advisor to the Government of India, New Delhi on "R & D Project Management".

SCHOLASTIC ACHIEVEMENTS

• Qualified in the Graduate Aptitude Test in Engineering (GATE), 2009 - 94.45%
• Recipient of Ministry of Human Resource and development fellowship for M.S. Students.