A Fashion Retailer that Sells Women’s Clothing has Created an IoT-enabled: Data Analysis and Mining Assignment, DBS

Assignment Details

A fashion retailer that sells women’s clothing has created an IoT-enabled store. This store has Bluetooth Low Energy (BLE) beacons installed on the ceiling. This way the retailer can track the customers’ in-store movements. In more detail, these movements are recorded as customers move through the store using a mobile app.

Using this infrastructure, the retailer has gathered a beacon dataset. This dataset is given in this assignment. The given dataset includes one excel file consisting of two excel sheets.

You can download this excel file from here. (The same dataset is available in the “Exam Assignment Part B” folder, as well).

• The first excel sheet represents the customers’ in-store movements. In more detail: customer_id is the unique customer identifier. Each time a customer visits the store, a session is created. This session includes the areas that the customer has visited. Each area has an ID. This ID is noted in Figure 1. The timestamp column indicates the time when the client entered a specific area. The timestamp is defined as the number of seconds elapsed since 1 January 1970 (UTC). For your convenience, the seq_id column shows the sequence in which a customer visited the areas within each session. Closing, each customer might have made many store visits (i.e. many sessions).
• The second excel sheet includes the area-id (as noted in Figure 1) and the garments it contains. For simplicity, each area includes only one garment category.

Apart from the beacons, the retailer has also placed RFID tags on each garment. Based on different RFID antennas placed in-the store, the retailer is able to track the garment in-store movements from an operational perspective. In more detail, the retailer has the following antennas:

• “Backroom RFID antenna”: is placed in the backroom entrance and tracks all the garments entering the backroom.
• “Replenishment RFID gates”: tracks the replenishment process and captures the garments that are moving from the backroom to the store floor and backward.
• “Entrance RFID gates”: track whether a customer tried to exit the store without purchasing a product.
• “Cashier RFID antennas”: track whether a product is purchased. Afterward, the RFID tag is deactivated.

Note: You are not able to track which customer carries which garment. You are able to track the garments solely in the areas having RFID antennas (i.e. backroom gate, replenishment gates, entrance gates, cashiers), as described above.

Based on the above you are asked to answer the following questions:

(A) Formulate one interesting data-driven question and answer it based on the beacon dataset. Structure and present your analysis based on the phases of the data analytics lifecycle.

(B) Regarding the last phase of the analytics lifecycle (i.e. operationalization), present and explain two examples showcasing two data-driven decisions your analysis results may support.

(C) How would you leverage the given beacon dataset and the RFID data that retailer gathers to support the company in the decision making, and create value to the customer and to the firm? Use examples, where appropriate, to explain your answer.

(D) What additional internal (i.e. owned by the retailer) data sources could you exploit and how could you help the retailer in creating more value? Use examples, where appropriate, to explain your answer.

(E) What other external datasets could you leverage to better understand and serve the retailer’s customers? Use examples, where appropriate, to explain your answer.