Faculty Directory by Area

title-arrow

Visiting Faculty

Nilotpal   Chakravarti
titile white arrow
Nilotpal Chakravarti
sub-icn

Operations Management, Quantitative Methods and Information Systems

deg-icn

  • PhD in Combinatorics & Optimization – University of Waterloo (1987)

Profile
  • Founder and Director of Smart Consulting Solutions Pte Ltd
  • 30 years’ experience in academia and industry
  • Taught at Northern Illinois University (1987-1989) & the Indian Institute of Management Calcutta (1989-1998)
  • Deep knowledge of operations research and statistical modeling techniques
  • Publications in prestigious refereed journals including Mathematics of Operations Research, Mathematical Programming and SIAM J on Optimization
  • Served in the Sales & Revenue Management Department of Singapore Airlines (SIA) where he trained staff, monitored the performance of Revenue Management (RM) systems and users, and led many major RM initiatives
  • Led projects in many domains including finance, airlines, logistics, media and government.
Publications
  • "Isotonic median regression; a linear programming approach", Mathematics of Operations Research, 1989, 14, 303-308.
  • "Active set algorithms for isotonic regression; a unifying framework", (with M. J. Best), Mathematical Programming, 47, 1990, 425-439.
  • "Sensitivity analysis in isotonic regression", Discrete Applied Mathematics, 45, 3, 1993, 183 -196.
  • "Some results concerning post-infeasibility analysis", European Journal of Operational Research, 73, 1994, 139-143.
  • "LP relaxation of the two dimensional knapsack problem with box and GUB constraints", (with A. Bagchi and N. Bhattacharyya), European Journal of Operational Research, 89, 1996, 609-617
  • "Calculation of Stability Radii of Combinatorial Optimization Problems", (with A. Wagelmans), Operations Research Letters, 23, 1998, 1-7.
  • "Minimization of Separable Convex Functions Subject to Chain Constraints", (with M. J. Best and V. Ubhaya), SIAM Journal on Optimization, V10, No. 3, 2000, 658-672.
  • "Easy Weighted majority Games", (with A. Goel and T. Sastry), Mathematical Social Sciences, 40, 2000, 227-235.
  • "Tea company steeped in OR”, OR/MS Today, April 2000, 32-34.
  • “Sensitivity Analysis for Shortest Path Problems and Maximum Capacity Path Problems in Undirected Graphs”, (with Ramkumar Ramaswamy and James B. Orlin), Mathematical Programming, Series A, 102, March 2005, 355-369.
Projects
Revenue Management
  • Development of an Origin-Destination level Revenue Management System at Singapore Airlines.
  • Various initiatives to improve Revenue Management within the Star Alliance.
  • Development of revenue management systems for two of India’s largest broadcasters.
  • Development of a revenue management system for the largest low-cost carrier in the Middle East.
  • Training SpiceJet staff in Revenue Management.
  • Conducting public training programs in Revenue Management in Bangkok, Dubai, Kuala Lumpur & Singapore.
Credit Scoring & Other Financial Services
  • Review and re-development of Credit Scoring models for an UK automotive loan provider.
  • Analyzing underwriting practices and making appropriate recommendations to an UK automotive loan provider.
  • Development of a Credit Scoring Model for the fifth largest US Bank to calculate the probability of default of credit card holders.
  • Development/validation of Credit Scoring Models for SME loans for two banks in Indonesia.
  • Development of a machine learning based scoring model to calculate the probability of a claim being fraudulent for a large Japan head-quartered insurance company in Singapore.
  • Modeling the behavior of customers of a major US auto finance company.
  • Data analysis to determine borrower segments in terms of profitability for a British financial institution specializing in mortgages.
  • Using data mining and operations research techniques for a British debt collection agency to determine the best matching of collectors to debts
  • Development of a tool for optimized portfolio allocation for a large Dutch pension fund. The tool uses methods such as genetic algorithms, simulated annealing and tabu search to maximize return while bounding risk. Risk may be measured as portfolio variance, semivariance or VaR. The tool accommodates many real-world constraints such as limits on sector, country, turnover, activity limit and beta, short sales and round lots.
Government & Big Data
  • Development of a Hadoop based Big Data system, for Singapore Government’s transport regulatory authority, to analyze daily transactions on buses to detect wrong fare deductions and emerging fault trends.
  • Development a scoring model to predict the child support payment behavior of non-custodial parents in several US states.
  • Development of software for optimal deployment of staff in Singapore schools for The Ministry of Education, Singapore. This is used to deploy around 4,000 – 5,000 staff to about 300 schools each year. The software matches officers to positions taking into account the requirements of schools, the qualifications of the officers and various systemic goals.
  • Forecasting future manpower requirement as part of development of software for manpower planning for the Ministry of Home Affairs Singapore.
Energy & Logistics
  • Planning off-shore drilling infrastructure over a 52-year planning horizon, for an European oil major.
  • Forecasting prices of ~60 petrochemicals in 3 geographies for a large multinational company.
  • Planning the inventory, storage and transport of ammunition for Singapore Armed Forces.
  • Development and implementation of a container loading software for a large packaging firm in Indonesia.
  • Development of software for planning the movement of empty containers for an Indonesian shipping company.
  • A study of the routes, frequencies and vessel assignments for a large Indonesian shipping corporation with a view to improving profitability.
  • A study of the logistics operations of Duncans Industries Ltd. (Packet Tea Division), one of the three largest packet tea companies of India. The study considered the entire supply chain for production/distribution of over 80 blends/lines of tea with several production centres, blending and packing units, and depots distributed all over the country. The results of the study were published in the April 2000 issue of OR/MS Today, the journal of INFORMS, the US Society for Operations Research/Management Science.
Other
  • Development of data analysis and optimization software for a market research company.
  • Development of software for targeted advertising and revenue maximization for one of the five largest US internet advertising companies.
  • Forecasting demand for Saudi telecom in post liberalization scenarios.
  • Developing scheduling algorithms to roster staff at the British Automobile Association.
  • Developing Media Optimization software for a ratings agency.