Estimation of Performance of Call Center Workers With Artificial Neural Networks Method
No Thumbnail Available
Date
2019
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Bilgi teknolojilerindeki hızlı gelişmeler inovasyonun da ötesine geçerek yıkıcı bir boyuta ulaşmıştır. Bu boyut yapay zekâ uygulamalarına yönelik geniş bir araştırma alanına yön vermiştir. Günümüzde yapay zekâ teknolojilerinin kullanımında önemli ölçüde ilerleme kaydedilmiştir. İnsana özgü olan biyolojik beyin sisteminden ilham alınarak Yapay Sinir Ağ (YSA) teknolojisi ortaya çıkarılmıştır. YSA insan beyninin fonksiyonlarından olan öğrenme yeteneğinin bilgisayarlara ve makinelere kazandırıldığı yapay zekâ teknolojilerinden olduğu, bu nedenle YSA'nın geçmişteki örnekleri öğrenme yoluyla kullanarak geleceğe ait yorumlar ve tahminler yapabilme özelliği gibi kuramsal çerçeve ele alınmıştır. Bu bakımdan YSA biyolojik sinir sisteminin matematiksel mimarideki modellemesidir. Tahmin performanslarının üstün olması YSA'lara birçok alanda başarılı bir şekilde kullanım alanları sağlamıştır. Bu çalışmada ise çağrı merkezi çalışanlarının sonraki aylara yönelik çalışma performansları YSA yardımı ile tahmin edilmiştir. YSA yöntem bilimi ile çağrı merkezlerinde çalışan müşteri/vatandaş temsilcilerinin performansı tahmin edilerek bulgular ve sonuçlar elde edilmeye çalışılmıştır. Çağrı merkezi sektöründeki firmaların başarılı olmasında ve yöneticilerin sağlıklı kararlar almasında temel koşul, neler yapılacağının önceden bilinmesidir. Bunun için öncelikle çağrı merkezlerinde çalışan personelin geçmiş aylara ait çalışma performans verileri saat olarak elde edilmiştir. Daha sonra tahmin amaçlı olarak kullanacağımız uygun YSA mimarisini oluşturmak için elde edilen mevcut veriler ile geri yayılım algoritması kullanılarak YSA eğitilmiş ve deneme yanılma yöntemiyle ağ parametreleri tespit edilmiştir. Nihai aşamada ise başarılı bir şekilde eğitilmiş ve test edilmiş olan bu uygun YSA'lar kullanılarak çağrı merkezlerinde çalışan personellerin daha sonraki aylara ait çalışma performansları başarılı bir şekilde tahmin edilerek uygun analiz ve değerlendirmeler yapılmıştır. Sonuç olarak bu çalışmada öngörü modellemesi tekniği olan YSA metodolojisi ele alınarak, çağrı merkezi sektöründe istihdam edilen çalışanların performanslarının tahmin edilmesine yönelik bulgularla sonuçlar başarılı bir şekilde elde edilmiştir. Elde edilen çıktılara göre bu çalışma göstermiştir ki çağrı merkezleri veya benzer sektörlerdeki yöneticilerin geleceğe dönük doğru ve sağlıklı kararlar alabilmelerini sağlayacak olan optimum çalışma şartlarının oluşturulması ve maksimum verimin alınabilmesi için çok ciddi imkanlar sağlamaktadır. Anahtar Kelimeler : Yapay Sinir Ağları, Çağrı Merkezleri, Tahmin Performansları, Öngörü Modellemesi, Optimizasyon
The rapid developments in information technologies have gone beyond innovation and reached a destructive dimension. This dimension has led to a wide research area for artificial intelligence applications. Today, significant progress has been made in the use of artificial intelligence technologies. Artificial Neural Network (ANN) technology has been unearthed by taking inspiration from human biological brain system. The theoretical framework such as the ability to make comments and predictions of the future by using the learning examples of the ANN as learning through the learning of the past, is discussed. In this respect, ANN is the modeling of biological nervous system in mathematical architecture. The superiority of the estimation performances has enabled successful use in many areas of ANNs. In this study, the performance of the call center employees for the following months is estimated with the help of ANN. With the help of ANN methodology, the performance of customer / citizen representatives working in call centers was estimated and the results and results were tried to be obtained. The basic condition for the success of the companies in the call center sector and the healthy decisions of the managers is to know what to do beforehand. For this purpose, working performance data of the personnel working in call centers were obtained as hours. Then, for the purpose of estimating the appropriate ANN using the data obtained with the back propagation algorithm to create the appropriate ANN data was trained and network parameters were determined by trial and error method. At the final stage, by using these appropriate NSAs, which have been successfully trained and tested, appropriate performance analyzes and evaluations have been made by estimating the working performances of the personnel working in call centers in the following months. As a result, in this study, the predictive modeling technique, ANN methodology has been discussed and the results obtained with the results of estimating the performance of the employees employed in the call center sector have been successfully obtained. According to the results obtained, this study showed that call centers or similar sector managers have the opportunity to create optimum working conditions that will enable them to make accurate and healthy decisions about the future and to get maximum efficiency. Key Words : Artificial Neural Networks, Call Centers, Forecasting Performance, Forecasting Modeling, Optimization. Quantity of Page : 86 Scientific Director : Doç. Dr. Remzi Tuntaş
The rapid developments in information technologies have gone beyond innovation and reached a destructive dimension. This dimension has led to a wide research area for artificial intelligence applications. Today, significant progress has been made in the use of artificial intelligence technologies. Artificial Neural Network (ANN) technology has been unearthed by taking inspiration from human biological brain system. The theoretical framework such as the ability to make comments and predictions of the future by using the learning examples of the ANN as learning through the learning of the past, is discussed. In this respect, ANN is the modeling of biological nervous system in mathematical architecture. The superiority of the estimation performances has enabled successful use in many areas of ANNs. In this study, the performance of the call center employees for the following months is estimated with the help of ANN. With the help of ANN methodology, the performance of customer / citizen representatives working in call centers was estimated and the results and results were tried to be obtained. The basic condition for the success of the companies in the call center sector and the healthy decisions of the managers is to know what to do beforehand. For this purpose, working performance data of the personnel working in call centers were obtained as hours. Then, for the purpose of estimating the appropriate ANN using the data obtained with the back propagation algorithm to create the appropriate ANN data was trained and network parameters were determined by trial and error method. At the final stage, by using these appropriate NSAs, which have been successfully trained and tested, appropriate performance analyzes and evaluations have been made by estimating the working performances of the personnel working in call centers in the following months. As a result, in this study, the predictive modeling technique, ANN methodology has been discussed and the results obtained with the results of estimating the performance of the employees employed in the call center sector have been successfully obtained. According to the results obtained, this study showed that call centers or similar sector managers have the opportunity to create optimum working conditions that will enable them to make accurate and healthy decisions about the future and to get maximum efficiency. Key Words : Artificial Neural Networks, Call Centers, Forecasting Performance, Forecasting Modeling, Optimization. Quantity of Page : 86 Scientific Director : Doç. Dr. Remzi Tuntaş
Description
Keywords
İşletme, Performans, Performans değerlendirme, Yapay sinir ağları, Çalışanlar, Çağrı merkezi, Business Administration, Performance, Performance evaluation, Artificial neural networks, Workers, Call center
Turkish CoHE Thesis Center URL
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
End Page
88