Search

© 2017 IEEE. Telecommunication industry is highly competitive, and mass marketing is not applicable anymore. Moreover, Mobile customers have different behaviors that urge telecom industries to differentiate their strategies to meet customers' needs. At the same time, mobile operators have an enormous amount of customer records, and data-driven approaches can help them to draw insights from this huge amount of data. Therefore, a data-driven segmentation approach can support marketing strategies to tailor their marketing plans. In this research, we adopt behavior and beneficial segmentation in a two-dimensional framework to segment customers. The results indicate that our method has an outstanding performance for customer segmentation. Moreover, we have recommended some marketing strategies based on each segment's behavior with the aim of increasing in Average Revenue Per User (ARPU) and decreasing in marketing expenses.

© 2017 IEEE. Telecommunication industry is highly competitive, and mass marketing is not applicable anymore. Moreover, Mobile customers have different behaviors that urge telecom industries to differentiate their strategies to meet customers' needs. At the same time, mobile operators have an enormous amount of customer records, and data-driven approaches can help them to draw insights from this huge amount of data. Therefore, a data-driven segmentation approach can support marketing strategies to tailor their marketing plans. In this research, we adopt behavior and beneficial segmentation in a two-dimensional framework to segment customers. The results indicate that our method has an outstanding performance for customer segmentation. Moreover, we have recommended some marketing strategies based on each segment's behavior with the aim of increasing in Average Revenue Per User (ARPU) and decreasing in marketing expenses.

Despite numerous experimental and numerical studies, fundamental understanding of how the matrix block height, the density difference between petroleum and gas, and matrix capillary pressure could affect the oil recovery from a single matrix block in naturally fractured reservoirs remains a topic of debate in the literature. In this work a one-dimensional gravity drainage model developed by Firoozabadi and Ishimoto (1994) is considered and numerically solved. The Fourier series method is applied for a numerical Laplace inversion of the dimensionless mathematical model; this type of inversion method has rarely been used in petroleum applications. The obtained results revealed that by increasing the matrix block height as well as the density difference between oil and gas, both the drainage flow rate and the cumulative production increased. In contrast, by increasing the matrix capillary pressure, both the drainage flow rate and the cumulative production decreased. The findings of this work can be helpful to better understand the behavior of the gravity drainage process in fractured porous media.

We have used the Maxwell–Stefan diffusion theory to model the mass transfer between tertiary-injected gas and residual oil blocked by water, in order to predict the time required for the rupture of the water barrier due to oil swelling. We have also designed and conducted a set of visualization micromodel experiments on various pure and multicomponent oil–gas systems to measure the water rupture time in tertiary gas injection processes. The experimental results show that the initial pressure and dimensions of the system, the oil and gas composition, and the gas solubility in water control the oil swelling process. The experimentally measured rupture times are then employed to evaluate the reliability of the model and to compare its accuracy with that of a similar one using classical Fick's law. Our modeling results show that both models are able to estimate the water rupture time for pure systems with an acceptable precision. As for multicomponent mixtures, however, only the Maxwell–Stefan theory is capable of modeling the molecular diffusion process correctly and yields values close to reality, while the use of Fick's law would lead to erroneous results. Deficiency of the latter model becomes more acute when the diffusion direction in reality is contrary to what the model indicates, which leads to failure in calculating any value for rupture time at all for these cases.