Publication Date

2018-07-28

Availability

Embargoed

Embargo Period

2020-07-27

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PHD)

Department

Management (Business)

Date of Defense

2018-06-28

First Committee Member

S.Y. Tang

Second Committee Member

Harihara P. Natarajan

Third Committee Member

Nan Yang

Fourth Committee Member

Joseph Johnson

Abstract

Rapidly developing technology is changing our life everyday. We are enjoying the benefits from the Information Age and exposed to more and more information. In my dissertation, I will discuss the value of information in the Economics and Supply Chain Management context, and investigate patient-level information in healthcare operations. In the first essay, I develop a Bayesian nested logit model which utilizes appointment confirmation data and estimates individual-level coefficients for patient-specific predictors. A log-likelihood comparison of model fit on twelve months of appointment data shows that the Bayesian model outperforms the standard logit model by about 30\% improvement in model fit. Additionally, the Bayesian model allows categorization of patients based on their appointment confirmation behavior. Finally, using patient-specific no-show probabilities as an input to a simulated appointment scheduler I find that the Bayesian model improves clinic profit over the standard logit model. In the second essay, I consider a supply chain where a manufacturer sources from an unreliable supplier. The supplier can take costly effort to improve its reliability, but has private information about its initial reliability level and improvement cost. I examine how these two types of asymmetric information affect manufacturer's profit, information rent, and channel loss differently. I aim to answer the following question: if the manufacturer were to learn the supplier's private information, which type of information is more valuable? The results indicate that the two types of information are equally valuable when the product value is low. When the product value is high, the reliability level information is more valuable. When the product value is medium, the reliability level information is more valuable when the degree of improvement is low, and the improvement cost information is more valuable when the degree of improvement is high.

Keywords

no-shows; Bayesian method; appointment reminder; supply disruption; supply reliability improvement; asymmetric information

Available for download on Monday, July 27, 2020

Share

COinS