USULAN PENGEMBANGAN PRODUK TABIR SURYA PT KOSMETIK MENGGUNAKAN DAIPA DENGAN MEMANFAATKAN ULASAN DARING
PT Kosmetik is an Indonesian FMCG company that offers a range of cosmetic products, including skincare and makeup. One of its popular products is sunscreen, which, despite its popularity, receives lower average ratings than other similar products on the Female Daily website. This suggests the com...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/83676 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | PT Kosmetik is an Indonesian FMCG company that offers a range of cosmetic products,
including skincare and makeup. One of its popular products is sunscreen, which, despite its
popularity, receives lower average ratings than other similar products on the Female Daily
website. This suggests the company may not have fully understood customer satisfaction. This
research uses dynamic asymmetric importance-performance analysis (DAIPA) to understand
how customer satisfaction may change over time and provide development recommendations
based on the customer satisfaction dimension (CSD). The formulation of DAIPA began with
the collection of online review texts on the Female Daily website by web scrapping for the ten
most popular sunscreen products and obtained around 59 thousand online reviews. Next, a
data preparation stage was conducted and a period of 5 years from 2019 to 2023 was
determined with two focal products (focal 1 and focal 2) and two competitor products (K1 and
K2). CSDs for sunscreen products were identified through latent Dirichlet allocation (LDA)
topic modeling and the CSDs were obtained, namely oiliness (MI), texture (TE), effectiveness
(EF), packaging (KE), affordability (HA), acne effect (JE), whitecast residue (WH), and after-
effect (EF). Review sentiments were identified using aspect-based sentiment analysis using the
bidirectional encoder representations from transformers (BERT) model. Next, the effect of
CSD fulfillment on customer satisfaction was measured using an artificial neural network
(ANN), specifically a Poisson deep neural network (PDNN). DAIPA plots were obtained, which
were then interpreted by each CSD according to DAIPA categories, namely high-performance
excitement (HE), high-performance performance (HP), high-performance basic (HB), low-
performance basic (LB), low-performance performance (LP), and low-performance excitement
(LE). Product development recommendations are formulated according to priority, starting
from the first priority, namely CSD "AE", the second priority by CSD "KE" and "JE", the third
priority by CSD "MI", the fourth priority by CSD "WH", and finally the fifth priority by CSD
"TE", "EF", and "HA".
|
|---|