Design and Implementation of a Marking Strategy to Increase The Contact Ability in The Call Centers Based on Machine Learning

Jamar is a company that belongs to the furniture sector, which manufactures and sells furniture and accessories for the home. Customer calls are one of the most trusted channels used in contact centers. Currently, the contactability indicator has a goal of 40% and is at 31%. The enemies of the efficiency of this channel are the terrible dimensioning, customers who evade answering these calls by identifying the numbers, non-market numbers in the databases, failures in the technological resources. Therefore, a proposal was made to design and implement a marking strategy in the call center, supported by a statistical model for dimensioning. Likewise, emerging technology such as Machine Learning is performed to help the marking strategy in outbound campaigns, reconfiguration of the dialplan to make it more efficient, and a redundant architecture design in the operators. Basic concepts of Teletraffic are explained, showing its primary functions, relevant for the management of the company's telephone system. In the same way, fundamentals of the Asterisk IP PBX are exposed, one of the most used in our environment due to its versatility and low implementation cost. The methodology of descriptive and applied research is used for the development of the project. The results and discussion show the dialing strategy and some call statistics from previous years, necessary to establish a correct dimensioning of the solution. The proposed solution allows having redundancy management for SIP and trunk operators, to have backup and reliability in case of failure. Keywords—network; calls; dialplan; machine learning; asterisk.

[1]  Villy Bæk Iversen,et al.  Teletraffic engineering and network planning , 2015 .

[2]  M. Kubát An Introduction to Machine Learning , 2017, Springer International Publishing.

[3]  I. Roque,et al.  METODOLOGÍA DE LA INVESTIGACIÓN , 2019, Cultura de emprendimiento corporativo en las micro, pequeñas y medianas empresas colombianas.

[4]  Rohit Kumar,et al.  Machine Learning—Basics , 2017 .

[5]  Brett Schwarz Asterisk open-source PBX system , 2004 .

[6]  Ioannis D. Moscholios,et al.  Teletraffic models beyond Erlang , 2014, 2014 ELEKTRO.

[7]  Neil Boyette,et al.  Towards a Low-Cost High-Quality Service Call Architecture , 2006, 2006 IEEE International Conference on Services Computing (SCC'06).

[8]  V. V. Maksimov,et al.  Development of interactive voice response models based on IP-PBX Asterisk , 2013, 2013 23rd International Crimean Conference "Microwave & Telecommunication Technology".

[9]  Yves Dallery,et al.  A flexible architecture for call centers with skill-based routing , 2012 .

[10]  Shai Ben-David,et al.  Understanding Machine Learning: From Theory to Algorithms , 2014 .

[11]  Jacek Ilow,et al.  Integration of Open Source and Enterprise IP PBXs , 2007, 2007 3rd International Conference on Testbeds and Research Infrastructure for the Development of Networks and Communities.

[12]  Ramjee Prasad,et al.  Distributed Redundancy or Cluster Solution? An Experimental Evaluation of Two Approaches for Dependable Mobile Internet Services , 2004, ISAS.

[13]  Dandan Yin,et al.  Computing on Base Station Behavior Using Erlang Measurement and Call Detail Record , 2015, IEEE Transactions on Emerging Topics in Computing.

[14]  Michel Mandjes,et al.  A large deviations approach to the transient of the Erlang loss model , 2001, Perform. Evaluation.

[15]  Mohammad Azam Khan,et al.  ASTERISK Based Open Source IP-PBX System for Accountable Customer Support Service , 2015, 2015 3rd International Symposium on Computational and Business Intelligence (ISCBI).

[16]  อนิรุธ สืบสิงห์,et al.  Data Mining Practical Machine Learning Tools and Techniques , 2014 .

[17]  Milos Orgon,et al.  Enterprise network with software Asterisk PBX based on the PLC technology , 2017 .