Search results for " Uncertainty"
showing 10 items of 777 documents
Approximate Bayesian Computation for Forecasting in Hydrological models
2018
Approximate Bayesian Computation (ABC) is a statistical tool for handling parameter inference in a range of challenging statistical problems, mostly characterized by an intractable likelihood function. In this paper, we focus on the application of ABC to hydrological models, not as a tool for parametric inference, but as a mechanism for generating probabilistic forecasts. This mechanism is referred as Approximate Bayesian Forecasting (ABF). The abcd water balance model is applied to a case study on Aipe river basin in Columbia to demonstrate the applicability of ABF. The predictivity of the ABF is compared with the predictivity of the MCMC algorithm. The results show that the ABF method as …
Latent Class Model to Test the Preferences Heterogeneity in the Perceived Information by Public Transport Users
2012
The aim of analysis is to understand as not reliable information influence the user behaviour and how much disincentive the public transport use. For this purpose, a Stated Preference Survey has been carried out in order to know the preferences of public transport users related to information needs and uncertainty on the information provided by Advanced Traveller Information System (ATIS). The perceived uncertainty is defined as the information accuracy. In our study, it was considered the difference between forecasted or scheduled waiting time at the bus stop and/or metro station provided by ATIS, and experienced one by user, to catch the bus and/or metro respectively. A questionnaire has …
Excessive vs. insufficient entry in spatial models: When product design and market size matter
2020
Abstract Under spatial product differentiation and product design, we identify conditions for either excessive or insufficient firm entry. We extend previous settings, based on the Salop circular model, to analyze the combined role of positive demand elasticity and endogenous targeted product design. First, we show that, given the number of firms, the equilibrium level of targeted design is either excessive or insufficient, depending on demand elasticity. Second, with free entry, we show that the degree of targeted product design increases with the relative market size and decreases with demand elasticity. Based on these effects, the interplay between demand elasticity and market size yield…
An evaluation of the environmental factors for supply chain strategy decisions using grey systems and composite indicators
2020
[EN] The purpose of this work is to assess the importance of environmental factors in a supply chain with four partners as a preliminary step to select the competitive strategies and objectives. To achieve this purpose, a real case study was carried out in a footwear supply chain, in which two approaches were used: the grey system theory and uncertainty analysis tools for composite indicators. In order to validate both approaches, a seven-phase research methodology was developed and applied to our case study. In addition, the priorization of environmental factors was calculated individually for each partner. The results allow managers to establish the competitive strategy that best suits th…
Examination of the least-squares method applied to the evaluation of physicochemical parameters with linearized equations
1989
Physicochemical parameters are frequently evaluated by the least-squares method after linearization of the equation which relates the experimental variables to the parameters of interest. When a given set of data is treated to evaluate (n+1) parameters. (n+2) nominally identical linearized equations are possible; each of these leads to different sets of values of the parameters which are also affected by different variances. These differences are also observed when statistical weights are used. Two methods for establishing which one of the (n+2) equations is best for fitting the data, i.e., simulation of experiments and propagation of errors, are discussed, compared and applied to a potenti…
Approaches to evaluate the virtual instrumentation measurement uncertainties
2002
This paper deals with the metrological characterization of virtual instruments. After a brief description of the features, the components and the working principle of the virtual instruments and the various uncertainty sources are analyzed. Then, two methods to evaluate the uncertainty of the measurement results are presented: a numerical method simulating the physical process of the A/D conversion, and an approximated theoretical method applying the "uncertainty propagation law" of the "guide to the expression of uncertainty in measurement." With both methods, the combined standard uncertainty of the measurement result is obtained, starting from the standard uncertainty generated by each s…
Measurement uncertainty estimation of a virtual instrument
2002
The law of uncertainty propagation is applied to a PC-based virtual instrument. According to the "guide to the expression of uncertainty in measurement", the measuring model and the various sources of uncertainty have been taken into account. The expressions of the combined output uncertainty in various basic digital signal processing blocks, are obtained starting from each input sample absolute accuracy, without taking into account the parameters which regard the overall behavior of an acquisition board, such as the effective number of bits. In order to verify this procedure, the results obtained by means of the theoretical analysis, are compared with the ones obtained from numerical simul…
Assessment of virtual instruments measurement uncertainty
2001
Abstract In this paper, two methods to evaluate the measurement uncertainty of a virtual instrument are presented: a numerical method simulating the physical process of the A/D conversion, and an approximated theoretical method applying the “uncertainty propagation law” of the “guide to the expression of uncertainty in measurement”. After a brief description of the features, the constitution and the working principle of the virtual instruments, the various uncertainty sources, are analyzed. With both methods, the combined standard uncertainty of the measurement result is obtained, starting from the standard uncertainty generated by each single source and without taking into account the para…
Propagation of errors due to incorrect positions of sources and detectors in wave-field tomography
2004
Tomographic data processed by 2D inversion programs can produce fairly large distortions due to incorrect source and/or detector positions. This problem is very serious in high-frequency electromagnetic tomography (GPR), due to the dimensions of the transmitter and receiver antennae. The errors can even be larger when coupled antennae are used (receiver and transmitter inside the same box) whose positions are not clearly known. Similar errors can be involved in seismic tomography, for instance when the mechanical connection between transducers and sample is defective. In this paper the problem has been studied using synthetic data which were calculated for different acquisition geometries. …
Data Treatment and Error Analysis in Kinetics
2000
Methods of data treatment in kinetics are reviewed from a practical viewpoint. The equations to describe the kinetics of chemical reactions and the methodology to obtain the models are presented. Moreover, the integration methods that are needed for the simulation of complex reactions are explained, from the simple first-order reaction to the complex general mechanism. However, it is essential to know the main sources of error in data treatment in order to reduce their effects on the results. Stress is put on the analysis of the errors due to the simplification of complex models. The different factors that influence the kinetics of the reaction are reviewed. How to take advantage of the kin…