Papers by J. George Shanthikumar
IISE Transactions, 2019
Simulation is considered as the most practical tool to estimate manufacturing system performance,... more Simulation is considered as the most practical tool to estimate manufacturing system performance, but it is slow in its execution. Analytical models are generally available to provide fast but biased estimates of the system performance. These two approaches are commonly used distinctly in a sequential approach, or one as alternative to the other, for assessing manufacturing system performance. This paper proposes a method to combine simulation experiments with analytical results in the same performance evaluation model. The method is based on kernel regression and allows considering more than one analytical methods. High-fidelity model is combined with low-fidelity models for manufacturing system performance evaluation. Multiple area-based low-fidelity models can be considered for the prediction. The numerical results show that the proposed method is able to identify the reliability of low-fidelity models in different areas and provide higher accurate estimates. Comparison with alternative approaches shows the method is more accurate in a studied manufacturing application.
Probability in the Engineering and Informational Sciences, 1998
In this paper we compare the variance of functions of random variables and functionals of point p... more In this paper we compare the variance of functions of random variables and functionals of point processes. Specifically we give sufficient conditions on two random variables X and Y under which the variances var f(X) and var f(Y) of the function f of these random variables can be compared. For example we show that if X is smaller than Y in the shifted-up mean residual life and in the usual stochastic order, then var f(X) ≤ var f(y) for all increasing convex functions f, whenever these variances are well defined. In the context of point processes we compare the variances var Φ(M) and var Φ(N) of the functional Φ of two point processes M = {M(t), t ≥ 0} and N = {N(t), t ≥ 0}. We provide sufficient conditions under which these variances can be compared. Specifically we consider comparisons between (i) two renewal processes and between (ii) two (homogeneous or nonhomogeneous) Poisson processes. For example we show that for any nonhomogeneous Poisson process N with a rate function bounde...
2021 Winter Simulation Conference (WSC), 2021
We describe a framework in which two parties have competing objectives in estimating the mean per... more We describe a framework in which two parties have competing objectives in estimating the mean performance of a stochastic system through sequential sampling. A "regulator" is interested in obtaining a correct measurement with an acceptable probability level, while the "stakeholder" in the project is interested in minimizing the cost associated with the output performance measure and the sampling cost. We demonstrate how the stakeholder can choose an optimal sampling rule to minimize their costs when they have private information which is not shared with the regulator. We further suggest how the regulator can choose a key controllable parameter of the optimal sampling rule in order to asymptotically approach a fair result for both parties.
In this study we develop a Decision Support System tool to monitor and understand the characteris... more In this study we develop a Decision Support System tool to monitor and understand the characteristics of emergency departments of NSW Hospitals management systems in detail to find cost effective solutions to service quality improvements. In the past researchers have used statistical, operational research or simulation techniques in finding solutions to improve management systems. In contrast to these procedures, we introduce a new way of integrating operations research and statistics to build an operations statistics model and simulate the system for forecasting. Statistical and data mining techniques are used to analyse data. Further computing and IT skills are used to create and integrate various functional modules in developing the Decision Support System tool. Finally the capabilities of the developed DSS tool will be illustrated by using several studies to improve efficiency of the Campbelltown Hospital emergency service facility.
Management Science, 2019
Inventory-based pricing under lost sales is an important yet notoriously challenging problem in t... more Inventory-based pricing under lost sales is an important yet notoriously challenging problem in the operations management literature. The vast existing literature on this problem focuses on identifying optimality conditions for a simple management policy while restricting to special classes of demand functions and to the special case of single-period or long-term stationary settings. In view of the existing developments, it seems unlikely to find general, easy-to-verify conditions for a tractable optimal policy in a possibly nonstationary environment. Instead, we take a different approach to tackle this problem. Specifically, we refine our analysis to a class of intuitively appealing policies, under which the price is decreasing in the postorder inventory level. Using properties of stochastic functions, we show that, under very general conditions on the stochastic demand function, the objective function is concave along such price paths, leading to a simple base stock list price pol...
Management Science, 2019
Please scroll down for article-it is on subsequent pages With 12,500 members from nearly 90 count... more Please scroll down for article-it is on subsequent pages With 12,500 members from nearly 90 countries, INFORMS is the largest international association of operations research (O.R.) and analytics professionals and students. INFORMS provides unique networking and learning opportunities for individual professionals, and organizations of all types and sizes, to better understand and use O.R. and analytics tools and methods to transform strategic visions and achieve better outcomes. For more information on INFORMS, its publications, membership, or meetings visit http://www.informs.org
Methodology and Computing in Applied Probability, 2018
Journal of the Operations Research Society of Japan, 1983
In this note, we derive bounds for the waiting time inasingle server queueing systern with genera... more In this note, we derive bounds for the waiting time inasingle server queueing systern with general independent arrival and general serviee times, using a regenerative process representation of this queue. We present easily computable new upper bounds for the variance and new upper and lower bounds for the second and higheT mornents ofthe waiting time, Also we present a sirnple approxirnation fbr the variance ofwaiting time.
Production and Operations Management, 2016
Subsidy programs are widely offered in both developing and developed countries to encourage consu... more Subsidy programs are widely offered in both developing and developed countries to encourage consumption of products that generate positive social, health, and environmental externalities. We study the effect of subsidies on product consumption under uncertain market demand. To reach a target consumer population, the program sponsor may subsidize a for‐profit or a not‐for‐profit firm on each unit of the product purchased by the firm or on each unit of the sale generated by the firm. We show that subsidy programs provide stronger incentives to a not‐for‐profit firm than to its for‐profit counterpart in inducing a large consumption whenever the sponsor is having a very limited budget or a very generous budget. When subsidizing a not‐for‐profit firm, the sponsor should always choose the purchase subsidy over the sales subsidy because the former can induce a larger consumption than the latter with the same subsidy spending. However, this is not always true when the subsidy program is adm...
______________________ 2. SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION/AVAILABILITY OF REPORT... more ______________________ 2. SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION/AVAILABILITY OF REPORT-__________________________________Approved for public release; distribution 2b. OECLASSIFICATION/DOWNGRAOING SCHEDULE unlimited. 4 PERFORMING ORGANIZATION REPORT NUMBER(S) 5. MONITORING ORGANIZATION REPORT NUMBER(S) AFOSRTR. 8 6-02 75 6i. NAME OF PERFORMING ORGANIZATION 5b. OFFICE SYMBOL 7s. NAME OF MONITORING ORGANIZATION
2016 Winter Simulation Conference (WSC), 2016
Simulation is widely used to predict the performance of complex systems. The main drawback of sim... more Simulation is widely used to predict the performance of complex systems. The main drawback of simulation is that it is slow in execution and the related compute experiments can be very expensive. On the other hand, analytical methods are used to rapidly provide performance estimates, but they are often approximate because of their restrictive assumptions. Recently, Extended Kernel Regression (EKR) has been proposed to combine simulation with analytical methods for reducing the computational effort. This paper has different purposes. Firstly, EKR is tested on different cases and compared with other techniques. Secondly, two different methods for calculation of confidence band are proposed. Numerical results show that the EKR method provides accurate predictions, particularly when the computational effort is low. Results also show that the performance of the two confidence band methods depends on the case analyzed. Thus, further studies are necessary to develop a robust method for confidence band calculation.
Proceedings. 1987 IEEE International Conference on Robotics and Automation
ABSTRACT
A I I E Transactions, 1980
The accuracy of the Content should not be relied upon and should be independently verified with p... more The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden.
Stochastic Processes and their Applications, 1987
A discrete time stochastic process {X,, n = 0, 1, 2,...} is said to be temporally convex (concave... more A discrete time stochastic process {X,, n = 0, 1, 2,...} is said to be temporally convex (concave) if E~b(X,,) is a nondecreasing convex (concave) function of n whenever ~k is a nondecreasing convex (concave) function. Similarly one can define temporal convexity and concavity for continuous time stochastic processes. In this paper we find conditions which imply that a given Markov process is temporally convex or concave. Some illustrative examples of stochastic temporal convexity and concavity in reliability theory, queueing theory, branching processes and record values are given. Finally an application of temporal stochastic concavity to a problem in computational probability is described.
We evaluate conditional value-at-risk (CVaR) as a risk measure in data-driven portfolio optimizat... more We evaluate conditional value-at-risk (CVaR) as a risk measure in data-driven portfolio optimization. We show that portfolios obtained by solving mean-CVaR and global minimum CVaR problems are unreliable due to estimation errors of CVaR and/or the mean, which are aggravated by optimization. This problem is exacerbated when the tail of the return distribution is made heavier. We conclude that CVaR, a coherent risk measure, is fragile in portfolio optimization due to estimation errors.
International Series in Operations Research & Management Science, 2003
... part collects articles on manufacturing systems modeling: John's persona... more ... part collects articles on manufacturing systems modeling: John's personal perspective on theevolution of the ... 2 STOCHASTIC MODELING AND OPTIMIZATION This volume is dedicated to honoring John Buzacott ... Friday," when the company fired a third of its head office staff, but ...
26th IEEE Conference on Decision and Control, 1987
ABSTRACT
Stochastic Models, 1986
Unclassified L. SECURITY CLASSIFICAiTION AUTHORITY 3. OISTRIBUTIONiAVAILABILITY OF REPORT NA Appr... more Unclassified L. SECURITY CLASSIFICAiTION AUTHORITY 3. OISTRIBUTIONiAVAILABILITY OF REPORT NA Approved for Public Release; Distribution a. OECLASSIFICATIONiOOWNGRAOING SCHEDULE Unl i mi ted NA * PERFORMING ORGANIZATION REPORT NUMBER(S) 5. MONITORING ORGANIZATION REPORT NUMBER(S) AF0SR-T 6-0 3 35 is. NAME OF PERFORMING ORGANIZATION 6. OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION
Communications in Statistics. Stochastic Models, 1986
Uni imi ted N/A 4 PERFCRmiNaG ORGANIZATICN REPORT NUMBERiSi 5. MONITORING RGANi .ATICN REPORT N4U... more Uni imi ted N/A 4 PERFCRmiNaG ORGANIZATICN REPORT NUMBERiSi 5. MONITORING RGANi .ATICN REPORT N4UiEAMSIS v*.OSR-TR. 8 A-o4 fav 6. NAME OF PERFORMING ORGANIZATION b. OFFICE SYMdOL :a. NAMt; ;F MUNITORIG .RGANIZAT.ON
Uploads
Papers by J. George Shanthikumar