New Rope Co., Ltd.
New Rope Co., Ltd. Released an AI trendbook that predicts “6 months ahead fashion trends” with 72% accuracy
Realized with fashion-specific image analysis technology and trend data accumulated over 4 years
New Rope Co., Ltd. has released an AI trend book that predicts fashion trends 3 months, 6 months, and 1 year ahead.
We will throw a new way out in the apparel industry, which is suffering from product disposal, large discounts, and missed sales due to uncertain demand.
The current accuracy is about 72%, and we will continue our research to further improve accuracy.
[Image 1
Many brands and retailers place orders for manufacturing about six months before actually selling products in stores or on e-commerce. For this reason, it is difficult to “hold only the appropriate inventory of items that are needed after ascertaining the trends of the world.” (Note 1)
This structural problem has resulted in over-ordering and
under-selling, which have long plagued the industry, weighing on both the environment and company profitability.
New Rope contributes by “predicting trends” while trying to solve this problem from various angles such as made-to-order production, shortening lead time, and off-price stores.
● Service summary
While general trend books are considered based on fashion collections, etc., this AI trend book consists of forecast data based on trend trends over a four-year period analyzed by SNS.
One of the features is that it incorporates various trends such as the Korean wave and skeletal diagnosis that are different from fashion collections.
Forecast data for 3 months, 6 months, and 1 year ahead can be used to optimize various processes such as merchandising, sales promotion planning, SNS operation, and customer service.
● Image analysis AI that enables trend analysis and accumulated trend information
New Rope has developed its own image analysis AI specialized for fashion. By continuing to manually create training data for nearly 10 years starting in 2014, we were able to automatically recognize 639 types of features, such as color, pattern, material, length, and collar shape. In addition, by analyzing images of fashion collections, SNS, EC, etc. with that AI, we have quantitatively analyzed fashion trends. [Image 2
Fashion items are made in small lots with a wide variety of products, and the short sales period makes it difficult to collect data, which is a problem unique to the industry.
Image recognition technology solves this problem. By classifying all fashion items by their characteristics, it enables comparison and analysis in the same data format across brands and seasons.
[Image 3
● High prediction accuracy based on accumulated data
This trend book predicts the future based on past trends in fashion. The accuracy is 72% for 6 months ahead and 78% for 3 months ahead. (Note 2) [Image 4
For each category such as tops, bottoms, dresses, bags, shoes, etc., predict the rate of change of features such as color, pattern, material, length, and collar shape.
AI Trendbook provides these predictions.
[Image 5
You can also intuitively see how popular trends are actually incorporated into the styling from snaps by popular Instagrammers. (Note 3)
It is expected to be used in a wide range of processes such as MD, sales promotion planning, SNS operation, and customer service. Ultimately, we aim to solve problems related to the environmental impact of inventory and the profitability of the industry.
[Image 6
This trend report can be purchased for 3,000 yen per copy by applying online. Please see the form below for details.
https://share.hsforms.com/14US9iqzdRF2xY1yyGLoGVA3wvwc
Note 1: The actual lead time varies depending on various factors such as the production background, item category, and fabric inventory. Note 2: Directional hit rate when classifying as flat when the volatility rate is around 30%, uptrend if it is above that, and downtrend if below. We made a prediction for December 2022 based on the data up to June 2022 and compared it with the actual value, and the same result was obtained.
Note 3: Snap is based on “#CBK” operated by New Rope.
● Related links
Apply for AI Trends Book: https://share.hsforms.com/14US9iqzdRF2xY1yyGLoGVA3wvwc “#CBK forecast” for trend analysis:
https://encyclopedia.cubki.jp/services/forecast/
Fashion AI “#CBK scnnr”: https://encyclopedia.cubki.jp/services/scnnr/ ● Company profile
Company name: New Rope Co., Ltd.
Address: Tsuzuki Gakuen Hatchery Shibuya, 14-1 Sakuragaoka-cho, Shibuya-ku, Tokyo
Representative: Satoshi Sakai, Representative Director
Business: Development and sales of fashion AI “#CBK scnnr”, media business URL: https://www.newrope.biz/
Details about this release:
https://prtimes.jp/main/html/rd/p/000000038.000024452.html
MAIL:cr@prtimes.co.jp