The Research Of Brand Apparel Assistant Planning System

As the research of independent apparel brand design and marketing are hot in domestic garment industry, the traditional way of apparel planning cannot meet the development requirements of diverse styles, small quantities and short-period. A scientifically effective management of apparel planning has become the key to the success of a garment enterprise.After learning the idea of overseas PLM products, this paper has designed an apparel design assistant system of our own brand garment industry which conforms to the current domestic planning status. The system could assist the director of fashion design to conduct fast and accurate apparel planning, which endow the garment products with brand characteristics. Furthermore, the system could reduce the risk of marketing by prospectively evaluation.The author's main work is summarized as below.1) SVG(Scalable Vector Graphics) which is a W3C recommendation is utilized as the format of the design draft, and a vector editor has been developed on the basis of Apache Batik Toolkit. Popular design patterns such as State pattern and Singleton pattern have benn implemented in order to make the code more liable to maintain and expand.2) As to the element process, we have resolved the problem of how to parse and edit basic elements and special elements. By the foundation of vector editor, function modules as style design module, fabric display module, and application of brand features elements module have been brought about.3) A model has been proposed, which demonstrates how visual identity elements are applied in some apparel design step. The model is to make the application of visual identity elements in apparel design more rapid and accurate, meanwhile helping the garment industry to shorten the design period and strengthen the brand image.4) According to the historical fashion color data, a new forecasting model based on the gray theory and modified BP neural network——G-LMBPNN is proposed. Firstly, the gray theory is applied to learn changes of historical data and obscure the data by accumulating. Secondly, Levenberg-Marquardt algorithm is utilized to train the BP network with samples made up of processed historical data. Finally, predicted result is attained from simulation and whitening. The research result proves that the G-LMBPNN model predict method is more precise than traditional gray model and possesses a faster convergence speed than the BP network.By system implement and model experiment, the main work of the paper proves effective in assisting the director to conduct scientificly and swiftly during brand apparel planning.