Search

The success of a development intervention is primarily defined by its longer-term results. Information and Communication Technology for Development (ICT4D) projects are no exception. In particular, the external context in which such projects are to be deployed plays a crucial role for achieving better development results. Thus, an analysis of contextual factors is essential for decision-makers at the project proposal screening and evaluation. Guided by the results-based management framework, a tool for influence assessment of contextual factors is designed and presented in this paper. The tool comprises a suggested model for common ICT4D contextual factors and a complementing method for their assessment. The method employs a scenario-based judgments elicitation process for data extraction together with Monte-Carlo simulation for data processing and aggregation. While the tool is not intended for comprehensive and deep analysis, the underlying streamlined process along with the easy to interpret assessment results arguably ensure further wide adoption. [ABSTRACT FROM AUTHOR]

Rising global temperatures have accelerated the water cycle, causing a redistribution of water resources that ranges from minor to extreme, arid areas are especially affected by this process due to the fragile nature of their eco-hydrology, which is more sensitive to climatic changes. The main effects of climate change to runoff are as follows: (1) The runoff in North Xinjiang has a strong positive relationship with precipitation, while that in the south slope of the Tianshan Mountains, the middle section of the north slope of the Tianshan Mountains and the Shule River has a strong positive relationship with air temperature; and there is a positive significant relation between summer runoff and 0 °C level height (FLH). (2) Human activities are presently the main driving forces behind runoff changes. In the Aksu, Kaidu, Shule and Heihe Rivers, the influence percentage of human activity on runoff is 90.4, 55.7, 63.1 and 78.8 %, respectively. (3) It was discovered that runoff in the Aksu and Yarkand Rivers is increasing with rising precipitation and temperature levels, and that runoff in the Hotan River is likewise increasing, but the rate of increase there is minimal. As runoff's response to temperature is more sensitive to precipitation changes, runoff in the Aksu and Yarkand Rivers will increase 1.4–7.0 % with every 1 °C rise in temperature. [ABSTRACT FROM AUTHOR]

The Extreme Learning Machine classifier is used to perform the perturbative method known as Sensitivity Analysis. The method returns a measure of class sensitivity per attribute. The results show a strong consistency for classifiers with different random input weights. In order to present the results obtained in an intuitive way, two forms of representation are proposed and contrasted against each other. The relevance of both attributes and classes is discussed. Class stability and the ease with which a pattern can be correctly classified are inferred from the results. The method can be used with any classifier that can be replicated with different random seeds. [ABSTRACT FROM AUTHOR]

A problem of parameter estimation for non-linear models may be solved using different approaches, but in general cases it can be always transformed to an optimization problem. In such a case the minimized objective function is a measure of the discrepancy between the model solution and available measurements. This paper presents the ADFIT program – a tool for numerical parameter estimation for models that contain systems of non-linear ordinary differential equations. The user of the program provides a model in a symbolic form and the experimental data. The program utilizes adjoint sensitivity analysis to speed up gradient calculation of the quadratic objective function. The adjoint system generating the gradient is created automatically based on the symbolic form of the model. A numerical example of parameter estimation for a mathematical model arising in biology is also presented. [ABSTRACT FROM AUTHOR]

In recent years, individual based epidemic simulations on synthetic social contact networks have been proved useful in supporting public health epidemiology. In use of such models, it is often desirable to understand how disease spreads through a small subset of the population; furthermore often data is available that allows us to develop more refined representations of social contact networks that span these sub-populations. Examples include: schools, office campuses, military bases, etc. Here, we develop a multi-level refinement and analysis technique to synthesize and embed subnetworks of an existing contact network when more detailed activity data about a small community is available. The methodology is illustrated by embedding detailed high school networks in a synthetic social contact network of the New River Valley, Virginia. [ABSTRACT FROM AUTHOR]

Public policy making has direct and indirect impacts on social behaviors. However, using system dynamics model alone to assess these impacts fails to consider the interactions among social elements, thus may produce doubtful conclusions. In this study, we examine the political science theory of greed and grievance in modeling civil conflicts. An agent-based model is built based on an existing rebellion model in Netlogo. The modifications and improvements in our model are elaborated. Several case studies are used to demonstrate the use of our model for investigating emergent phenomena and implications of governmental policies. [ABSTRACT FROM AUTHOR]

This chapter presents a resource optimization framework for wireless networks used for information exchange among the nodes of a distributed control system. In particular, the proposed framework achieves an optimal tradeoff between end-to-end delay robustness and network throughput for given delay thresholds imposed by the control application layer. To select an objective function for delay robustness, which can provide the desired tradeoff optimally, we employ sensitivity analysis. For maximizing the network throughput, an effective link transmission rate based power control problem is solved. The proposed resource optimization framework is then extended using an iterative suboptimal cross-layer algorithm to improve the link congestion fairness. Performance evaluation results show that a small compromise in the network throughput can provide a large delay robustness depending on the operating point. An improvement in the link congestion fairness performance at the cost of reduced network throughput is also studied. [ABSTRACT FROM AUTHOR]

In this paper, the relationship between the toll rate of roads and traffic volume assignment are analyzed based on traffic equilibrium assignment of different roads between an OD pair. And the application of sensitivity analysis is illustrated with a practical example, which shows the relationship between toll and traffic volume under the condition of competition among different sorting roads. [ABSTRACT FROM AUTHOR]

BP neural network as a data-mining technique can be used to factor analysis, but multi-layer BP neural network with hidden layer and between layers of neurons connected by weights staggered, so input variables on output variables. The size of impact is not intuitive. This research is based on different input variables change in value of the output variable degree of sensitivity analysis, the sensitivity of the size of the response by the input variables influence the output variables. Finally, application of the method on the cost factors affecting the sensitivity analysis, the factors that influence the ranking of a more reasonable level, the sensitivity of the BP neural network analysis of factors affecting the feasibility of certain. [ABSTRACT FROM AUTHOR]

Electrodynamic bearings exploit the eddy current effects that take place in a conductor rotating in a constant magnetic field to achieve passive levitation without the need of additional devices, as sensors, power electronics, etc. Nevertheless, the instability issues require a dynamic analysis that would guarantee the rotor stability. In the literature, electrodynamic bearings modeling is mainly based on quasi-static analysis, i.e., the study of the forces induced in a disk that rotates with a fixed eccentricity relative to the constant magnetic field. Although this approach is useful for a first-approximation analysis, it is intrinsically related to steady state conditions, and therefore limited. Besides, the quasi-static analysis is generally based on finite element simulations, in which geometries and material properties are tested through a trial-and-error method. In the following, a model of the bearing including a few parameters, but still able to describe both the quasi-static and the dynamic performance is presented. A sensitivity analysis based on the variation of the main design parameters of the electrodynamic bearing is performed. [ABSTRACT FROM AUTHOR]

In CPM network planning, one activity consumes float may affect other activities΄ floats, so that making the gross of network΄ floats change. The objective of sensitivity analysis on single activity to network floats is to measure the effect. Aiming at the matter, firstly, conception of eigenvalue parameter and method of computing the parameter are given; secondly, sensitivity of single activity to network total float, network safe float, network free float and network node float is analyzed by using the parameter; thirdly, functions of relation between quantity consumed by single activity and degree of network floats affected by the activity are designed, and correctness of the functions are proven; and finally, the feasibility is validated though example. [ABSTRACT FROM AUTHOR]

Non-urgent Emergency Department utilization has been attributed with increasing congestion in the flow and treatment of patients and, by extension, conditions the quality of care and profitability of the Emergency Department. Interventions designed to divert populations to more appropriate care may be cautiously received by operations managers due to uncertainty about the impact an adopted intervention may have on the two values of congestion and profitability. System Dynamics (SD) modeling and simulation may be used to measure the sensitivity of these two, often-competing, values of congestion and profitability and, thus, provide an additional layer of information designed to inform strategic decision making. [ABSTRACT FROM AUTHOR]

In this paper, we discuss sensitivity analysis in linear subspace method, especially on multiple-case diagnostics. Linear subspace method by Watanabe (1973) is a useful discriminant method in the field of pattern recognition. We have proposed its sensitivity analyses, with single-case diagnostics and multiple-case diagnostics with PCA. We propose a modified multiple-case diagnostics using clustering and discuss its effectiveness with numerical simulations. [ABSTRACT FROM AUTHOR]

We present rminer, our open source library for the R tool that facilitates the use of data mining (DM) algorithms, such as neural Networks (NNs) and support vector machines (SVMs), in classification and regression tasks. Tutorial examples with real-world problems (i.e. satellite image analysis and prediction of car prices) were used to demonstrate the rminer capabilities and NN/SVM advantages. Additional experiments were also held to test the rminer predictive capabilities, revealing competitive performances. [ABSTRACT FROM AUTHOR]

In this chapter, we present sensitivity analysis for performance measures of an irreducible perturbed Markov chain which is either discretetime or continuous-time. By using the UL- and LU-types of RG-factorizations, we can express the nth derivatives of the performance measures, including the stationary, transient and discounted cases. Furthermore, we apply the sensitivity analysis to study symmetric evolutionary games by perturbed birth death processes and asymmetric evolutionary games by perturbed QBD processes. [ABSTRACT FROM AUTHOR]

Combination failure of a solid waste landfill is one of main failure mechanisms. Based on the force limit equivalence theory, one formulae to calculate the factor of safety of the combination failure of landfill with a dam was derived in this article, and then the sensitivities of parameters to influence landfill stability were analyzed. The parameters with the greatest influence on combination failure are the landfill angle α, θ and β, followed by the internal friction angle of solid waste Φsw, then the interface friction angle of landfill liner system δ, and H2, H1,δD, B and ν are the last. From the point of stability and economy of landfills, the front angle α of landfill is most important. The compaction of solid waste could improve the internal friction angle and choosing of liner with high internal friction are conductive to the stability of landfills. [ABSTRACT FROM AUTHOR]

The research aims to explore the sensitivity of CRM indicators and to define specific conditions for their application according to customer historical information. The neural network analysis was applied for the research of classification task of distinguishing classes of potentially returning customers from those, who tend to leave the company. The research results revealed that application of traditional time and money-related variables have important advantage, as they are considered to make objective basis for judgement. The experimental research was performed by mining customer database of the travel agency. According to the experimental evaluation we explored that the classification model can not be uniformly applied throughout all the customer history during his lifecycle. The explored sensitivity of indicators suggests application of different neural networks fitting to the particular stages of the customer lifetime cycle. The neural network analysis results were summarized and the research insights presented, based on the dynamics of sensitivity of the customer indicators. [ABSTRACT FROM AUTHOR]

The characteristics of the hydraulic piping system in Tunnel Boring Machine (TBM) are large slenderness ratio, high fluid pressure and low flow velocity. With the work conditions of TBM, the beam model of long straight pipe was established. Under consideration of fluid structure interaction (FSI) and simple supported conditions, oscillation equation of piping system was solved by vibration formation decomposition. Having analyzed the effect of piping system parameters on the first order inherent frequency by the method of sensitivity analysis, pipe length was the most significant effective factor on inherent frequency of pipe system. Considering the relation between fluid pressure wave frequency and pipe inherent frequency, oscillation criteria based on the first order inherent frequency was put forward. Finally, optimized design was given for piping system design. [ABSTRACT FROM AUTHOR]

Quantitative multiattribute decision making methods are handy for making cooperative decisions, because the decision makers are interested in finding a common solution; even some times they have diverse goals. The initial data for cooperative decision making depend on the decision makers΄ own views, therefore the input data may be inconsistent. An important step in cooperative decision making is to perform a sensitivity analysis for the cooperative decision. In this paper we perform a sensitivity analysis for the cooperative decision with respect to the values of attributes and the values of the attribute significances. The judgements of the expert group are used for generating the values of attributes significances. Monte Carlo simulation is used for the generation of the attribute values. We propose to represent the final decision together with the results of sensitivity analysis and confidence level of the decision. We believe that this can help to increase the reliability of the cooperative decision making. [ABSTRACT FROM AUTHOR]

This paper discusses the use of the Automatic Differentiation approach in evaluating derivatives of functions represented by computer programs. We then considered a Cournot oligopoly modeled by a system of stochastic differential equations. The setting is that of a set of self-interested firms striving to adjust their productions in the direction of higher profits subject to mistakes or random shocks. The stochastic differential equations are solved by a numerical method and the profits are calculated using a Monte Carlo simulation. Then, Automatic Differentiation is used to propagate sensitivities along each path in an automated fashion. Numerical results have confirmed the intuition one may have that noisy environments can lead to important profit differences between firms as well as higher sensitivities as opposed to less noisy ones. [ABSTRACT FROM AUTHOR]

Practices for global sensitivity analysis of model output are described in a recent textbook (Saltelli et al., 2007). These include (i) variance based techniques for general use in modelling, (ii) the elementary effect method for factor screening for factors-rich models and (iii) Monte Carlo filtering. In the present work we try to put the practices into the context of their usage. We start by describing the present debate on the use of scientific models, and how uncertainty and sensitivity analysis can assist is testing model quality. We discuss Type I, II and III errors in the context of sensitivity analysis and what are the requirements for a good analysis. We also present sensitivity analysis in relation to post normal science (PNS) and model pedigrees. [ABSTRACT FROM AUTHOR]

Certification and quality assessment are crucial issues within the wine industry. Currently, wine quality is mostly assessed by physicochemical (e.g alcohol levels) and sensory (e.g. human expert evaluation) tests. In this paper, we propose a data mining approach to predict wine preferences that is based on easily available analytical tests at the certification step. A large dataset is considered with white vinho verde samples from the Minho region of Portugal. Wine quality is modeled under a regression approach, which preserves the order of the grades. Explanatory knowledge is given in terms of a sensitivity analysis, which measures the response changes when a given input variable is varied through its domain. Three regression techniques were applied, under a computationally efficient procedure that performs simultaneous variable and model selection and that is guided by the sensitivity analysis. The support vector machine achieved promising results, outperforming the multiple regression and neural network methods. Such model is useful for understanding how physicochemical tests affect the sensory preferences. Moreover, it can support the wine expert evaluations and ultimately improve the production. [ABSTRACT FROM AUTHOR]

Quantitative multicriteria evaluation methods may be effectively used in decision making by a cooperative group, because its members are seeking the solution of the problem by joint efforts. Solving multicriteria problems with conflicting aims, the cooperative group members behave like experts, making decisions based on their views and opinions [1]. The present paper considers the variation intervals of the criteria values and sensitivity of quantitative multicriteria evaluation methods to the initial data of multicriteria problem. Monte Carlo method is used for generating sets of initial data, and the analysis of sensitivity of multicriteria methods is made using a number of sensitivity analysis methods. The outcome of the analysis will be the development of methodology or an algorithm which could help to increase the reliability of decision making based on the use of quantitative multicriteria decision methods or their sets. [ABSTRACT FROM AUTHOR]

Measures programming is an important way for improving the effect of measures of oil fields. Based on the theory of fuzzy chance-constrained goal programming, a fuzzy chance-constrained goal programming model is put forward for measures programming of Oilfield, which uses the increasing production and cost as objectives. The process of its hybrid intelligent algorithm including fuzzy simulation, genetic algorithm and neural network is provided, and the sensitivity analysis on uncertain factors to determine the influence on the optimal result can also be carried on. Finally, an example is given to illustrate the validity of the model and approach. [ABSTRACT FROM AUTHOR]

Industry worldwide depends on Information and Communication Technology (ICT). Through large-scale blackouts of the public electricity supply telephone services and Internet connections are massively reduced in their functions, leading to cascading effects. Following analysis of selected, typical failure situations counter-measures to re-establish the public electricity supply in Austria to consumers are identified. This can serve also as an example for other countries. Based on the existing public electricity supply system, a sensitivity analysis both in power and in the ICT sector for the mobile and the fixed network is carried out. As a new possible solution ″smart grid″ or ″microgrids″ and the controlled operation of decentralized stable islands are investigated. [ABSTRACT FROM AUTHOR]

This paper presents research on the Vehicle Routing Problem (VRP) using a sweep heuristic method with 2-opt exchange and traveling salesman tours and an integer programming model for split delivery VRP model to select the best route to pick up and delivery customers from/to desired destination and depot. The modeling language, AMPL with CPLEX is used to develop the model and implement the sweep heuristic. The research found that the integer programming model produced the optimal result for some cases and failed to produce the optimal result for some cases. However, the sweep heuristic gave good solutions for all cases within a small amount of computational time. The research also investigated sensitivity analysis with respect to the vehicle capacity. The results indicate a savings in number of vehicle used with a small increase in vehicle capacity. The case study of University of The Thai Chamber of Commerce (UTCC) which provides bus services to pick up and deliver staff from/to home and university is selected to present in this paper. [ABSTRACT FROM AUTHOR]

In the hi-tech industries the emerging bottleneck(s) will restrict the whole production line and influence the factories' capability to make money, especially in Taiwan where OEM is the prevail way to earn low margin. In order to solve the problem efficiently, TOC can provide on-going improvement to the work sites. This paper shows how to use goal programming to identify the bottlenecks and the sensitivity analysis is presented. The analysis is illustrated with an empirical case. [ABSTRACT FROM AUTHOR]

The paper presents a framework for sensitivity analysis (SA) in spatial multiple criteria evaluation (S-MCE). The framework focuses on three aspects of S-MCE: spatiality, scope, and cardinality. Spatiality stresses the importance of spatial criteria and spatial weights that should be explicitly considered in GIS-based MCE. Scope relates to the extent of SA, ranging from local one-at-a-time criterion examination to global testing of interdependencies among the multiple criteria model components. Cardinality addresses the duality of motivation for performing SA, namely, single-user learning and group consensus building. The framework organizes the existing SA techniques according to spatiality and scope and can be used as a conceptual guide in selecting SA techniques fitting a task at hand. [ABSTRACT FROM AUTHOR]

In order to measure purchase preference on future electronic equipments, this paper proposes a method using Bayesian Network that considers causal relationships for questionnaire data. The proposed method is based on three layers network where the causal relationship is sequentially chained by personal attributes, senses of value and future services. Then, there is a problem that the probability value sensitivity becomes fewness when a model becomes complicated even if the evidence for same nodes is assigned. To overcome this problem and to perform probabilistic inference, this paper proposes stepwise assignment for a variety of evidence. As a result, the purchase preference such as ˵persons who make much of safety prefer products for an inventory management in their house″ is found. [ABSTRACT FROM AUTHOR]

The main performance variable in outrigger canoeing is boat velocity. The aims of this study were to derive a model of the mechanics of the outrigger canoeing (O1) stroke and to perform a sensitivity analysis to determine the key variables for increasing boat velo city. The model considered the fluid dynamic interaction between the paddle and water, the oscillatory motions of the paddler΄s upper body relative to the boat, the mass of the paddler and boat, and stroke rate. Due to the identical hulls between the O1 and K1 boats, measured hydrodynamic drag relationships for a K1 kayak were used. The aerodynamic drag acting on the paddler was assumed to be similar to that of a seated man. The model was solved using a fourth order Runge-Kutta solver with a fixed time step of 0.001s. The model was validated against the mean boat velocity for an elite female paddler over a 500m race distance. The paddler was filmed in order to determine the linear and angular trajectories of the paddle and the paddler΄s centre of mass through the stroke. A sensitivity analysis was then performed by increasing/decreasing each input variable by 1%. A difference of less than 3% was observed between modelled and measured mean boat velocity showing the model to be valid. The sensitivity analysis showed that the key variable influencing boat velocity was horizontal paddle velocity relative to the boat. The degree to which this variable can be increased is influenced by the strength of the paddler. The next most sensitive variables were the paddle΄s size and maximum drag coefficient. The magnitude of maximum drag coefficient is a function of the paddle design. Future research should now focus on determining the underlying mechanisms contributing to the increases in boat velocity elicited by changes to these key variables. [ABSTRACT FROM AUTHOR]

This paper presents an analysis of the Arruda accessory pathway localization method for patients suffering from Wolff-Parkinson-White syndrome, with modifications to increase the overall performance. The Arruda method was tested on a total of 79 cases, and 91.1 % localization performance was reached. After a deeper analysis of each decision point of the Arruda localization method, we considered that the lead aVF is not as relevant as other used leads (I, II, III, V1). The branch of the decision tree, which evaluates the left ventricle positions, was entirely replaced using different decision criteria based of the same biological parameters. The modified algorithm significantly improves the localization performance in the left ventricle. The overall performance reaches 94.9 %. A high localization performance of non-invasive methods is relevant because it can enlighten the necessary invasive interventions, and also reduces the discomfort caused to the patient. [ABSTRACT FROM AUTHOR]

In machine learning, the reliability estimates for individual predictions provide more information about individual prediction error than the average accuracy of predictive model (e.g. relative mean squared error). Such reliability estimates may represent a decisive information in the risk-sensitive applications of machine learning (e.g. medicine, engineering, business), where they enable the users to distinguish between better and worse predictions. In this paper, we compare the sensitivity-based reliability estimates, developed in our previous work, with four other approaches, proposed or inspired by the ideas from the related work. The results, obtained using 5 regression models, indicate the potentials for the usage of the sensitivity-based and the local modeling approach, especially with the regression trees. [ABSTRACT FROM AUTHOR]