On the use of entropy in Markovian model of rainfall

A statistical study of the annual rainfall data at Cocch Behar district during the periodt90 1 to t9S ~ has been undertaken by using. Markov chain model. One step 3 X3 Ma rcov chain model has beenused in (his study. The outcomes of the model reveal normal, bad and good year of rainfall a t the two stationsof thi s district. The hypoth esis of independence has been tested on the use of entropy and it has been ver ifiedusing likelih ood ra tio criterion. The results of the two methods are the same-tha t the yearly rainfall occurrencemay be regarded as independent at the two places o f the district.

A probabilistic study of monsoon daily rainfall at Calcutta by Markovian model and trend of rainfall during monsoon season

One-step transitional probability matrices are obtained by Markovian model from one transitional state to other for different rainfall characteristics (such as, non-rainy days, light rains, moderate rains, heavy rains and very heavy rains) at Calcutta (Alipore) for different monsoon months. The disorderness (unvertainty) of the transitional system of the monsoon rainfall characteristics are studied by Shannon’s model. The favourable or unfavourable condition of different states are also studied by redundancy test during monsoon period at Caluctta. A Long-run probability vectors of such rainfall characteristics are found out from probability matrices. A trend in rainfall amounts during monsoon period at Calcutta for each of the monsoon month, June to September, has also been studied here.

A Modeling Framework for Decision Support in Periprosthetic Joint Infection Treatment

In this paper, we present a framework, which aims at facilitating the choice of the best strategy related to the treatment of periprosthetic joint infection (PJI). The framework includes two models: a detailed non-Markovian model based on the decision tree approach, and a general Markov model, which captures the most essential states of a patient under treatment. The application of the framework is demonstrated on the dataset provided by Russian Scientific Research Institute of Traumatology and Orthopedics “R.R. Vreden”, which contains records of patients with PJI occurred after total hip arthroplasty. The methods of cost-effectiveness analysis of treatment strategies and forecasting of individual treatment outcomes depending on the selected strategy are discussed.

DATA-DRIVEN CREATIVITY: COMPUTATIONAL PROBLEM-EXPLORING IN ENGINEERING DESIGN

Creativity is required in engineering design. It is required in the aspects of problem-solving - conceptualizing a new solution to a problem, and problem-exploring - conceptualizing a new problem. Studies show that, in both aspects, creativity is a difficult task in practice. The aim of this study is to support the engineering design community by easing the difficulty in the problem-exploring practice. To achieve this, a computational problem-exploring (CPE) model is developed to mimic how design engineers identify a valid design problem. Consequently, a CPE tool - Pro-Explora V1 is developed based on the CPE model. The CPE model consists of a synergy of emergent computational technologies including data retrieval and machine learning. A Markovian model is employed in the CPE model to enable a data-driven random process for exploring design problems. In pilot test, Pro-Explora V1 generated some engineering design-related problems which are meaningful, unique, and could not be distinguished from naturally generated ones. It provides support to design engineers in problem-exploring at the early stage in engineering design. This study contributes to the global effort towards data-driven processes in the fourth industrial revolution.

Semi-Markovian approach to optimizing the time of systematic preventive maintenance and the level of maintenance over a finite horizon

This work is concerned with the problem of optimizing maintenance policies in terms of economical rewards and availability. We consider a system with multiple states in terms of healthy mode (good state, degraded state and failure state) and maintenance action (running state, stopped for maintenance). The level of maintenance (perfect or not) is also taken into account. We propose semi-Markovian model highlighting the effects of dwell times and transitions on economical rewards. We determine an optimal policy conditionally upon the current state according to eight decision parameters related to time intervals between two preventive maintenances and the level of maintenance. We show through a sensitive analysis that decision parameters have nonlocal effects that imply a multiple objective function. Hence, we propose a compromise by optimizing the asymptotic average reward.