This paper proposes a novel algorithm for optimizing decision variables with respect to an outcome variable of interest in complex problems, such as those arising from big data. The proposed algorithm builds on the notion of Markov blankets in Bayesian networks to alleviate the computational challenges associated with optimization tasks in complex datasets. Through a case study, we apply the algorithm to optimize medication prescriptions for diabetic patients, who have different characteristics, suffer from multiple comorbidities, and take multiple medications concurrently. In particular, we demonstrate how the optimal combination of diabetic medications can be found by examining the comparative effectiveness of the medications among similar patients. The case study is based on 5 years of data for 19,223 diabetic patients. Our results indicate that certain patient characteristics (e.g., clinical and demographic features) influence optimal treatment decisions. Among patients examined, monotherapy with metformin was the most common optimal medication decision. The results are consistent with the relevant clinical guidelines and reports in the medical literature. The proposed algorithm obviates the need for knowledge of the whole Bayesian network model, which can be very complex in big data problems. The procedure can be applied to any complex Bayesian network with numerous features, multiple decision variables, and a target variable.
In-Storage-Psychrophilic-Anaerobic-Digestion (ISPAD) is a sequentially fed batch treatment system operating at a temperature fluctuating with that of ambient. Because of its specific operation modes and the acclimation of its microbial groups, its microbial kinetics were determined from laboratory data, and a specific mathematical model was developed to simulate its process and to optimize its management. The objective of this study is therefore to validate this ISPAD model using further laboratory data obtained from batch tests conducted in flasks. For this purpose, glucose at 630 mg/L, was fed to 8-year-old ISPAD inoculum and digested at 18 °C. Changes in glucose, VFAs and pH were monitored along Environmental Management and Sustainable Development ISSN 2164-7682 2015 www.macrothink.org/emsd 194 with biogas production. The cross-validated coefficient of determination ( ) was used to determine the fit between the model prediction and the experimental values. The ISPAD model was able to strongly predict glucose degradation, VFAs, pH, and methane. However, the model weakly predicted the early CO 2 changes over time, likely because of its water solubility.
In-Storage-Psychrophilic-Anaerobic-Digestion (ISPAD) is a treatment system applicable to wastewaters stored for over 100 days, such as livestock wastes and municipal sludge. The ISPAD system differs from conventional reactors by being a sequentially fed batch process operating at a temperature fluctuating with ambient. The objective of this study was to develop a mathematical model to simulate the ISPAD process, verify the value of its microbial kinetics, and to simulate the pH evolution of its content along with its methane (CH 4 ) production. Furthermore, the values of the ISPAD microbial kinetics were compared to that of previous years to track for further acclimation to psychrophilic conditions. Simulation of ISPAD was achieved using the Simulink/Matlab software. The model was calibrated using laboratory data obtained from batch experiments using 7-year-old ISPAD inoculum, and glucose as substrate, and where glucose, VFAs and pH changes were monitored along with biogas production. The ISPAD model showed good agreement with the experimental data representing the system behaviour between 4 and 35 º C. Although microbial activity at 4 °C was much slower than that at 18 and 35 º C, it showed acclimation to low temperatures. Furthermore, comparison of microbial kinetic values over 3 years of field ISPAD monitoring demonstrated continued population acclimation, especially for the methanogens.
The three key drivers of a project success include cost, completion time, and scope, the interplay of which have a significant impact on the decision making in project management. In this study, we propose a theoretical framework to be used as a Project Management Decision Support System for understanding and balancing the interplay between the project cost and quality, which is a key component of the project scope. To this end, we develop a Decision Support Contract (DSC) for a project manager when outsourcing to a contractor whose delivery outcome is subject to quality risk. On the one hand, to reduce the risk of project failure, the contractor can invest in a quality improvement effort, the cost of which is the contractor's private information. On the other hand, the contractor's decision on quality improvement is unobservable to the project manager. In designing the DSC, we consider both problems resulting in information asymmetry between the project manager and the contractor. We first obtain the first-best solution assuming that the cost efficiency of the contractor is publicly known, and then solve for the second-best optimal cost plus incentive fee (CPIF) contract under information asymmetry. Our comparative study between the first-and second-best contracts reveals that the project manager may prefer to incur efficiency loss due to underinvestment decision by the high-cost contractor to reduce the information rent demanded by the low-cost contractor. Finally, we compare the effectiveness of CPIF contract to that of fixed-price contract, which enables us to characterize the value of incentive fee term for the project manager. This latter analysis reveals that incentive-fee term is more valuable when the improvement effort is more likely to reduce the quality failure risk.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.