Dr. Muhammad Yousaf Shad did Ph.D. from the University of Vienna, Austria. In 2006, He was selected as a YSSPer, IIASA (first researcher of Pakistan). He completed a two-year post-doctoral fellowship at NTNU Norway. His research interests are Stochastic Modelling and Optimization, Supply and Value Chain Design and Optimization, Network Optimization and Computing, Scenario Generation, and Credit Risk Modelling.
Dr. Muhammad Yousaf Shad's background includes Twenty- seven years of professional research/teaching experience, including 13 years as a faculty member at The Department of Statistics, Quaid-Azam University, Islamabad, and prior experience as a post-doctoral fellow (2 years) at the t Institute of Industrial Economics and Technology Management, NTNU, Norway and Ph.D. researcher (4 years) Institute of Decision Support System, University of Vienna, Austria. I also have a prior experience of eight years as a post-graduate class’s college teaching in Pakistan.
At Quaid-Azam University, Islamabad, NTNU Norway and University of Vienna, Austria, He gained exposure to research in:
1- Modeling in a stochastic environment
2- Applications, designs, and analysis in Bio-statistics fields.
3- Data collection, quantitative analysis, and results interpretation.
He mastered various blotting techniques; stochastic optimization, performed complex, ANOVA and ANCOVA; and become skilled in the use of:
Programming languages: including C, object oriented (C++), Visual Basic, mathematical (Matlab), modelling (Ampl, Zimple, GAMS), and database (PL/SQL) languages.
and Statistical Packages/language: Minitab, R, S-plus, SPSS, SAS.
He use to teach following subjects to undergraduate, graduate and PhD classes for the last 25 semesters: Operations Research, Stochastic Processes, Bio Statistics, Experimental Design, Regression Theory, Mechine Learning and Risk Management.
(Supply chain design and optimization )
(Stochastic Linear Optimization: A Case Study of Indus Basin Irrigation System )
University of Vienna, Austria
(Instability in Agriculture Production)
16-Sep-2010 - 28-Feb-2023
10-Feb-2008 - 10-Feb-2010
27-May-1995 - 15-Sep-2010
Empirical E-Bayesian estimation of hierarchical poisson and gamma model using scaled squared error loss function
The hierarchical models have not only a major concern with developing computational schemes but also assist in inferring the multi-parameter problems. The E-Bayesian is the expected Bayesian estimation that can be found by taking the integrals of Bayesian estimator using a hyper-prior with respect to the hyper-parameters. This study introduces the empirical E-Bayesian estimation that is coalesced with hierarchical modeling which prior to this has not been investigated. The scaled squared error loss function (SELF) has been used to estimate the parameter of Hierarchical Poisson-Gamma (HPG) model using empirical E-Bayesian estimation. The empirical E-Posterior risk is considered to be the evaluation standard. In addition, the consistency along with the asymptotic normality of the posterior distribution have been discussed. Furthermore, the empirical Bayes method is used to estimate the values of hyper-parameters via Maximum Likelihood (ML) method. The Monte Carlo simulation is executed to assess the precision of proposed estimators and a real-data application has been analyzed for illustration and comparison purposes.
On E-Bayesian analysis of the hierarchical normal and inversegamma model using different loss functions and itsapplication
This paper introduces the idea of hierarchical modelling to derive the variance (i.e. squared scale) parameter of the hierarchical normal and inverse gamma model using E-Bayesian estimation. We propose the idea of a hierarchical probability density function instead of the traditional hierarchical prior density function. We aim to derive E-Bayesian estimators with respect to the conjugate prior distribution IG(a,b) on the basis of the Balanced squared error loss function (BSELF) and Stein's loss function (STLF) for the unknown squared scale parameter. We use E-Posterior Risk as an evaluation standard. This study also intends to reveal the relationship among the E-Bayesian estimates under three distinct bivariate independent prior distributions of hyperparameters. The asymptotic properties of the E-Bayesian estimators are also evaluated. Monte Carlo simulations are prosecuted to study the efficiency of E-Bayesian estimators empirically and also, a real data application is analysed for exemplifying purposes.
A new fitness-based selection operator for genetic algorithms to maintain the equilibrium of selection pressure and population diversity.
A genetic algorithm is one of the best optimization techniques for solving complex nature optimization problems. Different selection schemes have been proposed in the literature to address the major weaknesses of GA i.e., premature convergence and low computational efficiency. This article proposed a new selection operator that provides a better trade-off between selection pressure and population diversity while considering the relative importance of each individual. The average accuracy of the proposed operator has been measured by χ2 goodness of fit test. It has been performed on two different populations to show its consistency. Also, its performance has been evaluated on fourteen
benchmark problems while comparing it with competing selection operators. Results show the effective performance in terms of two statistics i.e., less average and standard deviation values. Further, the performance indexes and the GA convergence show that the proposed operator takes better care of selection pressure and population diversity.
Soil conditioners improve rhizodegradation of aged petroleum hydrocarbons and enhance
the growth of Lolium multiflorum
Bioremediation and phytoremediation have demonstrated potential for decontamination of petroleum hydrocarbon-impacted soils. The total petroleum hydrocarbons (TPHs) are known to induce phytotoxicity, reduce water retention in soil, associate hydrophobic nature and contaminants’ in situ heterogeneous distribution, limit soil nutrient release and reduce soil aeration and compaction. The ageing of TPHs in contaminated soils further hinders the degradation process. Soil amendments can promote plant growth and enhance the TPH removal from contaminated aged soil.
The benefits of co-evolutionary Genetic Algorithms in voyage optimization.
Reducing emissions is of increasing global importance. Within shipping, the International Maritime Organisation’s regulations are putting pressure on companies to quickly reduce emissions. One solution is the optimisation of a ship’s route where even comparatively small reductions, in the order of 5%, provide sizeable cost and environmental benefits. The most recent advances from the Evolutionary Computation field have not been benchmarked on this problem, especially the co-evolutionary algorithms that provide the widest diversity of search. This paper compares state-of-the-art algorithms on three case studies, to show the impact of algorithm selection on the fuel consumption and expected voyage time. Four state-of-the-art Genetic Algorithms are selected to represent the leading families of Genetic Algorithm. The co-evolutionary approaches are shown to have the top performance, with cMLSGA (co-evolutionary Multi-Level Selection Genetic Algorithm) showing top performance on all the problems with the greatest potential reductions in fuel usage, 7.6% on average over the state of the art, and voyage times, 8.4% on average over the state of the art.
Sensitive proportion in ranked set sampling
This paper considers the concomitant-based rank set sampling (CRSS) for estimation of the sensitive proportion. It is shown that CRSS procedure provides an unbiased estimator of the population sensitive proportion, and it is always more precise than corresponding sample sensitive proportion (Warner SL (1965)) that based on simple random sampling (SRS) without increasing sampling cost. Additionally, a new estimator based on ratio method is introduced using CRSS protocol, preserving the respondent’s confidentiality through a randomizing device. The numerical results of these estimators are obtained by using numerical integration technique. An application to real data is also given to support the methods.
Interactive effect of biochar and compost with Poaceae and Fabaceae plants on remediation
of total petroleum hydrocarbons in crude oil contaminated soil.
The current study was dedicated to finding the effect of soil amendments (biochar and compost) on plants belonging to Poaceae and Fabaceae families. Plants selected for the phytoremediation experiment included wheat (Triticum aestivum), maize (Zea mays), white clover (Trifolium repens), alfalfa (Medicago sativa), and ryegrass (Lolium multiflorum). The physiological and microbial parameters of plants and soil were affected negatively by the 4 % TPHs soil contamination. The studied physiological parameters were fresh and dried biomass, root and shoot length, and chlorophyll content. Microbial parameters included root and shoot endophytic count. Soil parameters included rhizospheric CFUs and residual TPHs. Biochar with wheat, maize, and ryegrass (Fabaceae family) and compost with white clover and alfalfa (Poaceae family) improved plant growth parameters and showed better phytoremediation of TPHs. Among different plants, the highest TPH removal (68.5 %) was demonstrated by ryegrass with compost, followed by white clover with biochar (68 %). Without any soil amendment, ryegrass and alfalfa showed 59.55 and 35.21 % degradation of TPHs, respectively. Biochar and compost alone removed 27.24 % and 6.01 % TPHs, respectively. The interactive effect of soil amendment and plant type was also noted for studied parameters and TPHs degradation.
Seeking a balance between population diversity and premature convergence for real-coded genetic algorithms with crossover operator.
The major issue for optimization with genetic algorithms (GAs) is getting stuck on a local optimum or a low computation efficiency. In this research, we propose a new real-coded based crossover operator by using the Exponentiated Pareto distribution (EPX), which aims to preserve the two extremes. We used EPX with three the most reputed mutation operators: Makinen, Periaux and Toivanen mutation (MPTM), non uniform mutation (NUM) and power mutation (PM). The experimental results with eighteen well-known models depict that our proposed EPX operator performs better than the other competitive crossover operators. The comparison analysis is evaluated through mean, standard deviation and the performance index. Significance of EPX vs competitive is examined by performing the two-tailed t-test. Hence, the new crossover scheme appears to be significant as well as comparable to establish the crossing among parents for better offspring.
Estimation of population proportion using concomitant based ranked set sampling
This paper considers the concomitant based double ranked set sampling (CDRSS) for estimating the population proportion and compares with existing concomitant based ranked set sampling (CRSS) and simple random sampling (SRS) schemes. Moreover, taking into account information on a single concomitant variable, we also develop ratio-and exponential-type estimators, along with their biases and mean square errors (MSEs) up to first order of approximation, for precisely estimating the population proportion using CRSS and CDRSS schemes. The advantages of the ratio-and exponential-type estimators over SRS estimator are investigated in terms of the relative precision. A real data application is also given to support the theory.
Climate Change and dynamics in age-related Malaria incidence in South Africa: A focus on Zambia
In the last decade, many malaria-endemic countries, like Zambia, have achieved significant reductions in malaria incidence among children <5 years old but face ongoing challenges in achieving similar progress against malaria in older age groups. In parts of Zambia, changing climatic and environmental factors are among those suspectedly behind high malaria incidence. Changes and variations in these factors potentially interfere with intervention program effectiveness and alter the distribution and incidence patterns of malaria differentially between young children and the rest of the population.
Quantifying Media Effects, Its Content, and Role in Promoting
Background: Chikungunya is a vector-borne disease, mostly present in tropical and subtropical regions. The virus is spread by Ae. aegypti and Ae. albopictus mosquitos and symptoms include high fever to severe joint pain. Dhaka, Bangladesh, suffered an outbreak of chikungunya in 2017 lasting from April to September. With the goal of reducing cases, social media was at the forefront during this outbreak and educated the public about symptoms, prevention, and control of the virus. Popular web-based sources such as the top dailies in Bangladesh, local news outlets, and Facebook spread awareness of the outbreak. Objective: This study sought to investigate the role of social and mainstream media during the chikungunya epidemic.
A new Logistic distribution-based crossover operator for real-coded genetic algorithm
This paper proposed a new crossover operator called the Logistic crossover (LogX) which is used in conjunction with a well-known mutation operators Makinen, Periaux and Toivanen mutation (MPTM), non-uniform mutation (NUM) and power mutation (PM). The defined algorithm used the real encoded crossover and mutation operator. A set of 15 test problems have been taken from global optimization literature to test the performance of the proposed algorithm. Results are compared with some popular genetic algorithms (GAs) existing in the literature. The evaluation of performance of the proposed algorithm has been done by analysing the mean of the objective function values and by the Performance Index (PI). This comparative study shows that Logistic crossover operator (LogX) with three mutation operators outperform the other crossover operators.
- Stochastic Processes
- Probability an Statistics II
Facility location optimization
In this study, we presented a Maximum Coverage Location (MCL) model for emergency
services which is Rescue-15, Islamabad. We addressed this issue under fixed cost and
allocation of different emergency facility services. The MCL model is calibrated in two
Phases. Phase-I model marks the optimal facility sites that provide maximum coverage to all the
demand sites under budgetary constraints. The phase-II model solves the allocation problems of the
services to these facility sites that have been selected in Phase-I. Illustrative examples are given
to show how the proposed model can be used to optimize the locations of emergency facilities of
Rescue-15 Islamabad. We used General Algebraic Modeling System (GAMS) to solve these
Integrated design and operations of natural gas production and distribution under uncertainty.
The cost-efficient transportation of gas from the field
The project addresses optimization models and methodology to analyze how to build efficient
production and transportation systems for natural gas from wells to market. Both when it comes to
capacity expansions in the existing infrastructure, new field developments with infrastructure
connected to the existing infrastructure and the development of new areas. This project aims
towards lower costs and better capacity utilization. This will be achieved by focusing on
uncertainty management, having a value chain perspective and viewing the infrastructure system
as a whole rather than individual developments.
The optimization tools and models developed may be used to analyse the infrastructure in
Pakistan in order to optimize the production in the system over long time horizons. By having a
dynamic perspective over the years, a holistic perspective will be possible to analyse how
development of marginal fields and regional production may be integrated with the overall
production. We address enhanced recovery not at the field level by better technology, but rather at
the overall shelf level in a system perspective. This allows that developments, production and
transportation are phased in optimally and when possible. Better capacity utilization and lower
costs may increase the number of fields that are considered commercially sound to develop and
the possibilities to prolong regional production.
The project is also related to how to make optimal gas pressure available to the market because
analysis of different technology choices and investment alternatives may be analysed in a larger
perspective. Often several fields or wells in an area remote from the market may share some
common infrastructure like processing platforms, pipelines, LNG-plants or combinations of
available sources. Locations of infrastructure, timing of production, investments in compressors
etc. are example of decisions that will be possible to analyse and that will influence the
productivity and maintaining pressure in domestic utilization and industrial sector.