No abstract

No abstract

No abstract

I. Introduction T echnology revolution within the last decades makes the use of smart phones, phablets, tablets, computers in every instance of our daily and business lives. Finance sector and banks are also highly affected from this revolution and they adapt their systems to these new trends. More retail customers use digital banking channels with different layers of digital security and approvals result in nearly human free operation. On the other hand, commercial customers' behavior still tends to use bank branches for money transfer transactions due to the nature of their organizational structures. A customer could fax an instruction including money transfer transactions and authorized signature to the bank branch. This service is accepted as the customer to be situated in branch physically and promised to be commited in 90 minutes (SLA-Service Level Agreement). These instructions mostly consist of large count of transaction orders and also with money amounts much greater than the electronic transfer limits. Branch employees validate the signature on instruction, scan and then deliver it to the operation center. That's the way how branch employees spend much more time for customer relationships. Daily operation deals with this NP Type problem of planning the turnover and shifts according to banks standard operation procedures by forecasting the expected workload. Thus, proper workforce planning decreases the labor force costs whereas service quality and customer satisfaction increases. Call Centers and Operation Centers are the most common fields of this kind of businesses [1]. Banks aim to reduce the operational workload of branches through operation centers. Expert employees are appointed in operation centers to serve faster, more accurate and high quality service. Operation center employees digitalize the hard copy money transfer instruction and commit the transaction. Delivering operational transactions centrally provides service quality enhancement, improves customer satisfaction and saves serious amount of labor force. Currently, operation center managers usually predict the workload and assign the workforce manually depending on their previous experience of the team and operation leaders. Mostly, they reschedule and change their plans at the moment the workload density is observed. This type of delayed planning results in non-optimized working environment which should be avoided. Considering the hundreds of average employee numbers in operational and call centers this optimization cannot be ignored. Workforce optimization problem in bank operation centers is pretty similar to Call Center studies. Transaction volumes of the both business fields have dramatically changed [2]. Related work by authors already mentions predictability of transaction counts [3]. Moreover, formulas for inbound transaction volume are generalized by extracting time based attributes of historical data in recent works [4] [5]. Similarly, some other studies including smoothing methods would be adapted to the problem [6]. Also, thes...

ATMs are physical interaction points between financial institutions and real customers. Storing physical cash causes renouncing to get interested. On the other hand, customer satisfaction requires to store the necessary cash amount. This concern becomes even more critical for countries having high-interest rate and overnight interest rates are higher. In this paper, we will show that daily cash withdrawals are predictable and we will propose a cost function for replenishment optimization. Experiments show that proposed model decrease idle balance dramatically.