Intelligent manufacturing challenges and directions of the baijiu starter culture-daqu industry: Microbiome and engineering perspectives

Abstract

Background

Chinese Baijiu, a popular fermented alcoholic beverage cherished for its cultural role in socializing, entertaining, and collecting, heavily relies on Daqu as a pivotal starter culture that defines the flavor profile and quality of Baijiu. The process of Daqu fermentation involves raw material preparation, Daqu block molding, and incubation, traditionally fermentation through empirical knowledge and manual techniques, leading to erratic product quality. The intelligent transformation and enhancement of the Daqu industry are currently exploratory, constrained by unclear microbiome mechanisms influenced by biotic factors and engineering dynamics governed by abiotic factors.

Scope and approach

This review summarizes the fermentation characteristics, microecology, and abiotic and biotic factors affecting the quality of different Daqu types, discusses the progress and challenges of sensor data acquisition, machine learning-assisted data analysis, proportional-integral-derivative (PID) control system, and intelligent equipment in the Daqu intelligence application, and proposes future research directions integrating ecological and engineering mechanisms. It aims to provide novel insights into the application of intelligent technologies for Daqu production in Baijiu industry, and improve the current industrial practices in data tracking, process control, and quality assurance of Daqu intelligent transformation.

Key findings and conclusions

Ecological and engineering factors combine to determine the quality of the Daqu. Heat and moisture movement and conduction are the primary factors contributing to the spatial heterogeneity of Daqu. The standardization of raw materials and processes, orthogonalization of microbiomes, along modularization of the abiotic influence mechanisms form the foundation for intelligent engineering of Daqu. Promising technologies such as multi-modal heterogeneous algorithm configuration tools hold potential for advancing intelligent control within the Daqu industry.