SEARCH RESULT

In the competitive business world of our present time, the importance of using information by interpreting it correctly is increasing. Businesses use the data they own to differentiate, come to the forefront and to become more successful and to have more profits. The fast consumption sector, which is a field in which data management is valuable and which yields fast results, requires that retailers become able to analyze effectively what conditions consumers demand in terms of achieving competitive advantages (Geuens et al., 2003). In our present time, the methods, which are applied by retail managers in display planning, focus on sales, profitability and competition. These methods cause that businesses take steps that do not make any progress and that are not accurate as well as yielding easy and fast results and are valid according to the limited viewpoint. Many well-known incorrect trends push some product groups backwards rather than increasing sales. It causes that the product groups that have the potential are out of fame without becoming famous. The main aim is to sell the products which the consumer does not consider buying, and which are not in the mind of the consumer; and by so-doing, to increase the sales with the products that have less circulation. The development of new products should not be neglected by merely focusing on the targeted sales rates and/or profitability. In this respect, the purpose of the present study was to determine the consumer behaviors about which fast consumption products are preferred together, to investigate the agreement between crossproduct display and consumer behaviors, and to observe the effect of these on sales in the departments intended for promoting sales. In the present study, the Market Basket Analysis was performed with the Association Rules and Apriori Algorithm, which are among data mining models. The Association Rules search and find the relations between data, and explain the connection between each data and other data. The Association Rules are among the most important techniques of data mining. The weekly sales data of the customers of a large retail chain-store in Istanbul Anatolian side were employed in the present study. The data set of 9 main groups and 48 subgroups were examined over a weekly period with 6.502 activities of customers who made multiple product choices from among 7.232 transactions with the help of the SPSS Modeler 18.0 Program. The Cross-Product Analysis togetherness was made one-byone by changing the premise number as 1 in examining the cross products. By considering the observed product togetherness, a new proposal list was made instead of the existing crossproduct list. The proposed new display plan was shared with the businesses, and was ensured that it was applied at the store for one week, and the effect it is on the sales was observed. As a result of the study, it was determined that the proposed product display list increased the sales at a rate of 60% when compared to the current exhibition list. Thirteen new product groups, which were not included in the current list, were included in the proposed list; and more than half of the increase in the sales was covered by the new product groups. In other words, the consumers responded to the new product groups. It was also observed that the consumer preferences had a transition towards healthy snacks. For example, it was observed that the consumer group who purchased baby products preferred healthy bar products. It was also observed that different product groups like pasta and sauces, canned foods and pickles, soap and shower-bath products were preferred together as well as different categories which complemented each other were preferred together. It is expected that the results of the present study might show that the Association Rules/Basket Analysis, which are among data mining models in the retail sector, can yield meaningful and effective results. It is also expected that the present study will contribute to the productivity and to the efficient placement of goods on shelves, more effective campaign planning, and even with stock management. Günümüz rekabetçi iş dünyasında bilginin doğru yorumlanarak kullanımasının önemi giderek artmaktadır. İşletmeler farklılaşmak, öne çıkmak, daha başarılı ve kârlı haline gelmek amacı ile sahip oldukları verileri kullanmaktadır. Veri yönetiminin değerli olduğu ve hızlı sonuç verdiği bir alan olan hızlı tüketim sektörü perakendecilerin rekabet avantajları elde etmeleri açısından tüketicilerin hangi koşullarda ne istediklerini efektif olarak analiz edip farkında olmalarının gerektirmektedir(Geuens ve diğ., 2003). Günümüzde perakende yöneticilerinin teşhir planlamasında uyguladıkları yöntemler satış, kârlılık ve rekabet odaklıdır. Bu yöntemler kolay ve hızlı sonuç vermelerinin ve kısıtlı bir bakış açısına göre geçerli olmalarının yanı sıra yerinde sayan ve sağlıklı olmayan kararlar alınmasına sebep olmaktadır. Pek çok doğru bilinen yanlış eğilimler satışı arttırmak yerine bazı ürün gruplarını daha da geriye itmektedir. Potansiyeli olan ürün gruplarının parlayamadan sönmesine sebebiyet vermektedir. Asıl amaç, tüketicinin almayı düşünmediği, aklında olmayan ürünleri sattırmak ve bu sayede sirkülasyonu az olan ürünlerle satış artışı sağlamaktı. Sadece hedeflenen satış oranına ve/veya karlılığa odaklanarak yeni ürünlerin gelişimi ihmal edilmemelidir. Bu bağlamda çalışmanın amacı, hızlı tüketim ürünlerinin hangilerinin birlikte tercih edildiğine yönelik tüketici davranışlarının tespit edilmesi, satışı arttırmaya yönelik reyonlar da uygulanan çapraz ürün diziliminin tüketici davranışları ile uyumunun araştırılması ve satışa etkisini gözlenmesi amaçlanmıştır. Çalışmada veri madenciliği modellerinden Birliktelik Kuralları ve Apriori algoritması ile sepet analizi yapılmıştır. Birliktelik kuralları, veriler arasındaki ilişkileri arayıp bulur ve her verinin diğerleri ile olan bağlantısını açıklar. Bu çalışma da İstanbul Anado

Business and Organization Research (International Conference)
BOR

Necati Cem Dümrek Tuğba Kıral Özkan