International Research Journal of Applied and Basic Sciences © 2014 Available online at www.irjabs.com ISSN 2251-838X / Vol, 8 (10): 1516-1519 Science Explorer Publications Using fuzzy logic (type II) in the intelligent ATMs’ cash management Mojtaba Zandevakili, Mehdi Javanmard Tehran payame noor University Corresponding author: mojtaba zandevakili ABSTRACT: The mass of the issues facing today's ATMs, cash management optimization can help banks to manage the system dynamically. The main question in this optimization is, forecasting demand for cash from the ATM, at specified time intervals (e.g., daily or weekly or monthly). Since the structure of the signal (the predicted demand for cash) is nonlinear with many factors, including: days, weeks, days, months, holidays, and even the location of the ATM is connected, the common methods of identification is usually not a good answer and should be used soft computing techniques such as fuzzy logic and neural networks. In this study, the trend methods used in this field explores the application of fuzzy logic (type II) in this issue has been dealt with. Keywords: ATM, cash management, artificial neural networks, fuzzy logic (Type II), soft computing. INTRODUCTION Today, ATMs are one of the most immediate banking system hardware. Optimal management of cash and the availability of ATM network is one of the most important factors in business services. Use efficient and effective cash management can help banks to dynamically system management (Darwish, 2013). Recently most banks are changing their eyes towards obtaining higher performance in the cash management in ATM (Simutis R., Dilijonas, Bastina, Friman, & Drobinov, 2007) . The question is to determine the optimum amount of cash that each branch in the specified time periods (e.g., daily) should leave the ATM. Note that ATM may charge extra deposit pocket money to meet the interest of the bank and the deposit is better to be zero. Therefore, the amount of such charge shall be determined by the minimum subsequent investment money we have left. Since the structure of the signal (demand to withdraw money from an ATM) is nonlinear and depend on many factors, such as day of week, day of month, holidays, and even on a busy street or recreational facility where the ATM is located at the weekend, usually conventional identification methods haven’t a good response and should be used soft computing techniques such as fuzzy logic and neural networks. Here's an overview of solutions and methods that have been proposed to solve this problem, it is. Since most of the proposed methods use neural networks and fuzzy logic, they will be listed, in the next section. Artificial neural networks and fuzzy logic Neural networks are computational systems and techniques which simulation platform aimed at remembering information and learning functions of the human brain. In this networks if a cell is damaged, other cells can compensate for its absence and also be involved in its reconstruction. These networks are capable of learning. An adaptive learning system is done by using the examples, by the new inputs, synapse weights are changed so that the system will produce the right answer. If we consider a graph with an equivalent network, network training process is to determine the weight of each edge and the biases. The artificial neuron can be defined, in this case the number of input and output of each neuron has a bias, and each entry has a weight. Although artificial neural networks aren’t comparable with natural nervous system but their features preferred them where there are needs to learn a linear or nonlinear mapping, like as: the ability to learn, the scattering data, the generalization capability, parallel processing and robustness. In short, fuzzy logic can be said that although the words and concepts such as hot, cold, tall, short, young, old, etc. do not point to specific and detailed, but everyone understands surprising flexibility of mind and use them in decisions and conclusions. Variable in nature or in the calculation are two kinds: quantitative values that can be expressed with a certain number and qualitative values that be expressed based on