Browsing by Subject "Stochastic reward nets"

Decisions in health care must often be made under inherent uncertainty; from treating patients, to provisioning medical devices, to operational decisions at an outpatient clinic. The outcomes depend on the health of patients as well as the availability of health care professionals and resources. Complex models of clinic performance allow for experiments with new schedules and resource levels without the time, cost, unfeasibility, or risk of testing new policies in real clinics. Model-based methods quantify the effect of various uncertain factors such as the availability of personnel on health care quality indicators like patient wait times in a clinic.

Despite their purported value, few opportunities have existed to test models from data collection through optimization. This dissertation develops a clinic model from end-to-end, beginning with a description of the medical practice, to data collection, to model validation, to optimization. Specialty medical practice is abstracted into treatment steps, measured electronically, and verified through systematic observation. These data are anonymized and made available for researchers. A validation framework uses the data to develop and test candidate models, selecting one that maximizes predictive accuracy while retaining interpretability and reproducibility. The resulting model is used in improving schedules via heuristic optimization. Clustering the results reveals clinic performance groups that represent different goals in clinic quality.

A blockchain is an immutable record of transactions (called ledger ) between a distributed set of mutually untrusting peers. Although blockchain networks provide tremendous benefits, there are concerns about whether their performance would be a hindrance to its adoption. Our research is focused on Hyperledger Fabric (HLF), which is an open-source implementation of a distributed ledger platform for running smart contracts in a modular architecture. This thesis presents our research on performance modeling of Hyperledger Fabric using a Stochastic Petri Nets modeling formalism known as Stochastic Reward Nets (SRN). We capture the key system operations and complex interactions between them. We focus on two different releases of HLF, viz. v0.6 and v1.0+ (V1). HLF v0.6 follows a traditional state-machine replication architecture followed by many other blockchain platforms, whereas HLF V1 follows a novel execute-order-validate architecture. We parameterize and validate our models with data collected from a real-world Fabric network setup. Our models provide a quantitative framework that helps compare different deployment configurations of Fabric and make design trade-off decisions. It also enables us to compute performance for a system with proposed architectural improvements before they are implemented. From our analysis, we recommend design improvements along with the estimates of performance improvement. Overall, our models provide a stepping stone to the Hyperledger Fabric community towards achieving optimal performance of Fabric in the real-world deployments.