Customers Clustering Based On Rfm Score Using Genetic Algorithm  

Oleh:

Purnomo, Muhammad Ridwan Andi ; Fajarwati, Nur Riana

Jenis: Article from Proceeding
Dalam koleksi: PROCEEDING The 8th International Seminar on Industrial Engineering and Management (8th ISIEM). Atria Hotel & Conference, Malang, Indonesia March 17 – 19, 2015, page IM-66.
Topik: customers loyalty; customers relationship management; RFM; clustering; genetic algorithm
Fulltext: Paper 61 Muhammad Ridwan Andi Purnomo - Islamic University of Indonesia.pdf (432.03KB)

Isi artikel

Today, companies should select the right marketing strategy for their customers for being competitive in turbulent environment. It is required for keeping existing potential customers as the main profits resource. There are several methods usually used to analyse customers loyalty. Most of conventional methods usually just consider frequency of product consumption to analyse customers characteristics. However, such methods have lack in analysis of money spent by customers and time between arrival of customers. Actually, customers who came frequently in a period of time and customers who came rarely but spent so much money could not be neglected in the customers characteristic analysis. RFM considers these 3 factors, consumption interval (R), frequency (F) and money amount (M) in analysing customers loyalty. In this research, raw data about customers transaction would be analysed using RFM technique. Further, clustering technique would be applied in order to group customers who have similar characteristics. The clustering was conducted using Genetic Algorithm (GA), and further it called as GA-Clustering. The clustering analysis is supervised clustering and the number of cluster was determined based on number of CRM strategy that have been determined in advance through discussion with firm’s owner. The result of this research is customers cluster based on RFM value and suitable CRM strategy that must be applied to every customers cluster.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali