UCLA

The dissertation consists of three chapters that study topics in information chains. Information chains acquire, process and distribute content (information or data) in a way analogous to physical goods in supply chains. However information chains are different in that demand does not deplete inventories. Rather, information can lose value over time, become erroneous, or false, so that it needs to be purged. Furthermore, it is often not possible to plan or control the production of content, and the arrival of data elements or content can be stochastic. In effect flow in the chain is driven by content arrival, rather than inventory depletion. In the first chapter, we describe the flow of information in the content processing and storage portion of the chain using different models, and formulate decision problems related to capacity planning. The objectives include the size of the content data base (as a proxy for value) and the time required for content to be available to users.

In the second chapter, we consider a dynamic capacity planning problem for information content management in supply chains, where information is processed to be entered into a data base, which is then available to customers and users. Information processing is done by workers, and the capacity decision requires the determination of the number of workers required to process a time varying work load. The average time taken for processing is an important performance parameter, and this drives capacity decisions. In the short term planning problem it is possible to vary capacity by using overtime and part-time work. The problem is formulated as a multi-period, nonlinear, mixed integer program. Clearing functions are used to capture processing delays. Small problems can be solved optimally, but large cases can be challenging. We develop Lagrangean relaxation methods to decompose the problem and generate lower bounds, and propose heuristics for solving the problem efficiently.

The third chapter models and studies a decision problem faced by supplier of information content. With customers that are sensitive to both price and release time of content, we study profit maximization problem for a monopolist provider. Then we look at the case of supplier with two downstream distribution channels where the channels have to compete on setting their price and release time for the identical content to attract more customers. When supplier fixes the release time for the content and charges fixed prices to both channels, we find that there is an equilibrium pair of prices for the channels to charge the customers.