Development of Protocol for Computerized XRF-ED Zari Testing Using Artificial Intelligence Technique

Abstract

The fine gold and silver zari thread is an integral part of the silk handloom weaving predominantly practiced in the southern part of the Indian peninsula. The zari thread consists of a silk core, wrapped with flattened silver-copper alloy wire and electroplated with gold. In a brocade silk saree, a minimum of 240 g (1 Marc) of zari is used along with 540 g of dyed silk to produce a silk saree of 6 m in length. The cost of zari is up to 60% of the price of the raw material needed to produce the saree. Earlier, the silver purity of zari thread was assessed using the volumetric analysis technique. Recently, due to the sharp rise in the price of precious metals, the gravimetric technique, which is more time-consuming, has been adopted to test all the zari ingredients as a content percentage to help the fair-trade practice of zari thread. The advent of the computerized XRF-ED technique has accelerated the process of estimating the metal purity values instantly. It could not be used as a tool to estimate the cost of the zari thread as it is traded based on the concentration of silk. Thus, this research aimed to develop the computerized XRF-ED zari testing protocol using the artificial intelligence technique to assess the zari content values. The research findings revealed that the results of the zari thread constituents as content percentage estimated using AI technique are comparable and at par with the test results of the gravimetric analysis. Thus, the XRF-ED testing can be popularized in the field to assess the quality of zari used in the handloom silk saree available to consumers. Findings could also benefit the overall development of the handloom industry faced with tough competition from the silk saree woven with zari imitation thread.