Sep24-29_News_Vendor - Single-Period Stochastic Decision...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Single-Period Stochastic Decision Models The News Vender Problem Other One-period Models Reference Single Period Stochastic Inventory Model: News Vendor Problem Siqian Shen 1 1 Dept. of Industrial and Systems Engineering University of Florida Fall 2009 1 / 24 Single-Period Stochastic Decision Models The News Vender Problem Other One-period Models Reference Outline 1 Single-Period Stochastic Decision Models 2 The News Vender Problem 3 Other One-period Models 4 Reference 2 / 24 Single-Period Stochastic Decision Models The News Vender Problem Other One-period Models Reference Stochastic Demands and Evaluation Methods Recall that we assumed the demand is a known constant in our deterministic EOQ model. However, this assumption is usually unrealistic. Demand usually behaves randomly following a specific distribution in real life. Assume that demand D is an integer-valued discrete random variable with probability P ( D = d ) = p ( d ) . For instance, P ( D = 40 ) = p ( 40 ) = . 5 and P ( D = 60 ) = p ( 60 ) = . 5 means there is a “50-50" chance that the next day’s demand will be 40 or 60. 3 / 24 Single-Period Stochastic Decision Models The News Vender Problem Other One-period Models Reference Stochastic Demands and Evaluation Methods We evaluate our decision of order size q by using the expected value E ( q ) = X d p ( d ) TC ( q , d ) TC ( q , d ) is the total annual cost given order size q and demand realization d (for the stochastic case, we have a pool of values that might become a realization with corresponding probability of p ( d ) ). We also assume that our decision q only made once under this assumption (i.e., single-period decision models). 4 / 24 Single-Period Stochastic Decision Models The News Vender Problem Other One-period Models Reference Marginal Analysis Given a demand distribution, how to find the q that minimize E ( q ) ? Definition and Illustration of convex functions : In plain words, convex function is something that if you try to connect any two points on the function by using a line segment, the function itself will be always beneath the line. 5 / 24 Single-Period Stochastic Decision Models The News Vender Problem Other One-period Models Reference Marginal Analysis Given a demand distribution, how to find the q that minimize E ( q ) ? Definition and Illustration of convex functions : In plain words, convex function is something that if you try to connect any two points on the function by using a line segment, the function itself will be always beneath the line. 6 / 24 Single-Period Stochastic Decision Models The News Vender Problem Other One-period Models Reference Marginal Analysis Interpreted by our problem, the total cost will decrease as we increase the order size q , and then start increase beyond a “point of no return"....
View Full Document

This note was uploaded on 03/01/2010 for the course FFDFD dssss taught by Professor Aabb during the Spring '10 term at Abraham Baldwin Agricultural College.

Page1 / 24

Sep24-29_News_Vendor - Single-Period Stochastic Decision...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online