APPLYING MONTE CARLO SIMULATION AND RISK ANALYSIS TO THE FACILITY LOCATION PROBLEM

Facility location enigmas have become a more strategic decision than just finding the lowest require to be paid [i]or[/i] undergone space to house a company's harvest When choosing the placement of a distribution center a company must weigh the of recent origin freight costs and the take away from of a new or leased arrangement against its necessary service horizontals as well as several other factors. Companies can also save upon logistics costs in choosing the correct location by means of incorporating the inherent risk and variability that are involved in a facility location decision. However, in reality, greatest in quantity companies do not analytically consider risk and variability in choosing a location. This article at hands a methodology to help determine candidate locations and then management a financial risk analysis to determine the ideal location of a fresh facility.

INTRODUCTION

A public problem facing companies is the issue of require to be paid [i]or[/i] undergone versus service. A rapidly growing company will notice that service horizontals decrease and transportation costs increase without an ongoing thorough analysis of the changes to their existing transportation network.

This article focuses upon finding the ideal location of a fresh warehouse or distribution center. To disentangle this problem, several models exist to aid in conducting a network analysis. This article describes by what mode to utilize a family of facility location types known as "P-median models" in determining potential candidate locations. However, in practice, any well built location original may be used in determining these locations. The effort of this article is dedicated to enhancing these prototypes by constructing a financial risk analysis for the propos candidate positions. from one side this analysis, different costs associated with different geographic locations are considered. For greatest in quantity variables within the financial analysis, Monte Carlo simulation techniques are applied to account for the risk in the creation or utilization of a facility. [i]or[/i] part of to the other the use of financial analytical manners and Monte Carlo simulation, this article compares the alternatives discovered from one side the P-median model and provides an ideal solution.

REVIEW OF LITERATURE

Economic changes that be met with during a facility's lifetime can drastically alter the appeal of a particular site, turning today's optimal location into tomorrow's investment mistake Determining the best locations for of recent origin facilities is thus an important strategic challenge (Owen & Daskin, 1998) This decision can also impact various parts of a company's fill up chain and can have a long-term result on total logistics costs. Therefore, the advantages of using tools like simulation and risk analysis in facility location can minimize the financial risk involved. In new years, research has started to focus upon the stochastic aspects of facility location modeling. greatest in quantity notably, there has been work in facility location decisions utilizing reliability types (Snyder & Daskin, 2003) and inventory queuing modeling (Nozick & Turnquist, 1998) These methodologies combine various facility location protoplasts with various analytical techniques to determine ideal locations. However, these archetypes do not utilize the combination of simulation and risk analysis techniques to determine these locations. This approach is fresh to facility location literature as it utilizes Monte Carlo simulation and risk analysis to determine the best "economic-risk" location [i]or[/i] part of to the other the computation of the awaited annual worth. This article outlines a methodology that accounts for issues that can affect the overall economic benefit (eg transportation require to be paid [i]or[/i] undergones regional gas prices, labor issues). After the determination of candidates, this approach can be used by the agency of spreadsheet users with a financial background. In the past, this emblem of modeling has been almost completely execut by means of operations research professionals.

THE FACILITY LOCATION PROBLEM

Company ABC has seen tremendous extension in product demand over the past 2 years and is now looking for the ideal placement of another distribution center from one side the P-median model, possible candidate location solutions can be generated. The objective of this P-median pattern is to minimize the weighted distance from the storehouse (distribution center candidate) to all customer sites assigned to that magazine The weights used here are in the form of physical weights of annual shipments associated with a customer site. The type formulation (Panchalavarupu, 2000) can be seen in Appendix A.

Several software packages exist to aid an engineer familiar with linear programming in creating and solving these designs For this problem, AMPL software with a CPLEX solver was utilized to generate candidates.

This P-median design was analyzed with 30(??) locations from Company ABC's 297 in every one's mouth customer locations. The company's popular distribution centers were included in this analysis, on the contrary the current locations could not be altered.

Typically, this impressed sign of model would start with 30(p) locations and work down toward the required amount of distribution center However, in this instance, we were looking for a sufficient amount of West Coast candidate locations outside of the rife distribution center location of sees Angeles, California. Company ABC management felt that on-time delivery to its West Coast customers was not at the same horizontal as its eastern customers and the company felt another West Coast distribution center in a different location was exigencyed to counter this effect.