This article formulates a model for finding the optimal delivery time performance guarantee. The expected profit model is solved to find a closed-form expression for the optimal delivery time promise. The simple, yet powerful model gives new insights into performance service guarantees in general and delivery time guarantees in particular. Many manufacturing and distribution firms guarantee to meet a standard delivery time promise and pay significant compensation to customers when deliveries are late. If the firm promises its customers a delivery time that is too short, it will frequently not make the promise, have to pay significant compensation, and possibly lose market share over time. If the delivery time guarantee is too long, customers will find the delivery time unattractive and will buy elsewhere. Hart (1993) and others (Hill 1995) call this a “performance service guarantee.” Although a delivery time performance-guarantee scenario will be used as the context for this article, other performance service guarantee contexts could have been used as well. Other similar performance service guarantees include no-stockout guarantees (Hart 1993), waiting time guarantees (Friedman and Friedman 1997; Kumar, Kalwani, and Dada 1997), and up-time maintenance guarantees (Hill 1992). It should be noted, however, that an unconditional satisfaction guarantee (Hart 1988) is more complex and is not addressed in this article. According to data collected by the Center for Advanced Purchasing Studies, delivery promises are far from perfect in many industries in the United States. On-time delivery benchmarks for several industries are summarized in

The purpose of this paper is twofold. First, it serves to describe a new strategy, called Structured Database Monte Carlo (SDMC), for efficient Monte Carlo simulation. Its second aim is to show how this approach can be used for efficient pricing of path-dependent options via simulation. We use efficient simulation of a sample of path-dependent options to illustrate the application of SDMC. Extensions to other path-dependent options are straightforward. 1

INTRODUCTION Simulation is a technique for numerically estimating the performance of a complex stochastic system when analytic solution is not feasible [LaKe91, Sc90]. This section discusses both discrete-event and Monte Carlo simulation techniques. In discrete-event simulation models, the passage of time plays a key role, as changes to the state of the system occur only at certain points in simulated time. Queueing and inventory systems can be studied by discrete-event simulation models. Monte Carlo simulation models, on the other hand, do not require the passage of time. Monte Carlo simulation models have been used in estimating eigenvalues, estimating ß, and estimating the quantiles of a mathematically intractable test statistic in hypothesis testing. Simulation has been described [BrEtal87] as "driving a model of a system with suitable inputs and observing the corresponding outputs." Accordingly, the following three subsections d

In analyzing the output process generated by a steady-state simulation, we often seek to estimate the expected value of the output. The sample mean based on a finite sample of size n is usually the estimator of choice for the steady-state mean; and a measure of the sample mean's precision is the variance parameter, i.e., the limiting value of the sample size multiplied by the variance of the sample mean as n becomes large. This paper establishes asymptotic properties of the conventional batch-means (BM) estimator of the variance parameter as both the batch size and the number of batches become large. In particular, we show that the BM variance estimator is asymptotically unbiased and convergent in mean square. We also provide asymptotic expressions for the variance of the BM variance estimator. Exact and empirical examples illustrate our findings. Authors' addresses: Chiahon Chien, Synopsys, Inc., Mountainview, CA 94043, U.S.A., chien@Synopsys.com; David Goldsman, School of Industrial...

this paper is the development of statistical estimation procedures for fitting the rate and mean-value functions of an NHPP. In particular, we have developed a least squares procedure for estimating the parameters of an NHPP with an EPTMP-type rate function. This procedure uses a least squares method to fit the mean-value function. In addition, we have developed a weighted least squares formulation of this problem, along with a diagnostic analysis of why weighted least squares fails in problems with certain first- and second-order moment structures such as that arising in the estimation of the mean-value function of an NHPP. Furthermore, we have conducted a comprehensive experimental performance evaluation the least squares procedure. The results of these experiments as well as our experience in using these procedures to fit the mean-value function of an NHPP to actual data such as the arrival processes of organ donors and patients, indicates that these procedures are capable of adequately modeling a large class of arrival process encountered in practice

A simulation model is successful if it leads to policy action, i.e., if it is implemented. Studies show that for a model to be implemented, it must have good correspondence with the mental model of the system held by the user of the model. The user must feel confident that the simulation model corresponds to this mental model. An understanding of how the model works is required. Simulation models for implementation must be developed step by step, starting with a simple model, the simulation prototype. After this has been explained to the user, a more detailed model can be developed on the basis of feedback from the user. Software for simulation prototyping is discussed, e.g., with regard to the ease with which models and output can be explained and the speed with which small models can be written.

The model used in this report focuses on the analysis of ship waiting statistics and stock fluctuations under different arrival processes. However, the basic outline is the same: central to both models are a jetty and accompanying tankfarm facilities belonging to a new chemical plant in the Port of Rotterdam. Both the supply of raw materials and the export of finished products occur through ships loading and unloading at the jetty. Since disruptions in the plants production process are very expensive, buffer stock is needed to allow for variations in ship arrivals and overseas exports through large ships. Ports provide jetty facilities for ships to load and unload their cargo. Since ship delays are costly, terminal operators attempt to minimize their number and duration. Here, simulation has proved to be a very suitable tool. However, in port simulation models, the impact of the arrival process of ships on the model outcomes tends to be underestimated. This article considers three arrival processes: stock-controlled, equidistant per ship type, and Poisson. We assess how their deployment in a port simulation model, based on data from a real case study, affects the efficiency of the loading and unloading process. Poisson, which is the chosen arrival process in many client-oriented simulations, actually performs worst in terms of both ship delays and required storage capacity. Stock-controlled arrivals perform best with regard to ship delays and required storage capacity. In the case study two types of arrival processes were considered. The first type are the so-called stock-controlled arrivals, i.e., ship arrivals are scheduled in such a way, that a base stock level is maintained in the tanks. Given a base stock level of a raw material or ...

Multiple generators are often required in simulation studies, for instance, to facilitate synchronization for variance reduction purposes, and multiple independent streams per generator are helpful to make independent replications. A portable set of software tools is described for uniform random variates generation. It provides for multiple generators running simultaneously, and each generator has its sequence of numbers partitioned into many long (disjoint) sub streams. Simple procedure calls allow the user to make any generator “jump ” ahead to the beginning of its next sub stream, back to the beginning of its current substream, or back to the beginning of its first substream. A simple switch permits a change from regular to antithetic variates or vice versa. Implementation issues are discussed An efficient and portable code is also provided for computing (as MOD m) for any positive integer values of a < m, s < m, and m <2 b-1 on a b-bit computer. This code is used to implement the package. A Pascal implementation for 32-bit computers is described. The basic underlying generator for this implementation has been proposed in a previous paper; it combines two multiplicative linear congruential generators and has a period of p = 23 x 1018

We present an alternative simulation run allocation scheme for maximizing efficiency in simulation experiments for decision making under uncertainty. The issue of simulation efficiency is addressed from two perspectives: i) We want to minimize the total computation cost, with a constraint that the overall simulation quality must be higher than a desired level; ii) we would like to maximize the simulation quality with the constraint that the total computation cost can not exceed a given budget. While these two problems look different, we show that the solutions to these two problems are identical. Comparisons with other procedures show that our approach can achieve a speedup factor of 3~4 for a 10-design example. The speedup factor is even higher for problems having a larger number of designs.

Much of the rare-event simulation literature is concerned with the development of asymptotically optimal algorithms. Because of the difficulties associated with applying these ideas to complex models, this paper focuses on sub-optimal procedures that can be shown to be much more efficient than conventional crude Monte Carlo. We provide two such examples, one based on “repeated acceptance/rejection ” as a mean of computing tail probabilities for hitting time random variables and the other based on filtered conditional Monte Carlo. 1