在现代统计过程控制中，越来越多的实际应用中包含多个属性变量，且每个属性变量包括两个或多个属性水平，这种多元属性数据经常用一个多维列联表来刻画。为了更好地监控多维列联表数据，本项目主要开展三个方面的研究：. 1）将Group Lasso方法应用到log-linear模型中去，提出一些新的统计量来构建稀疏列联表数据下的多元属性控制图；. 2）将变量的顺序信息应用到一个扩展的log-linear模型中去，对含有有序变量的列联表数据提出一些新的监控方案；. 3）基于高阶马氏链方法和背景树方法，发展一些观测值自相关时多维列联表数据下的控制图方案。

In modern Statistical Process Control (SPC), more and more real applications involve multiple categorical quality characteristics and each characteristic has two or more attribute levels, which can be displayed by a multi-way contingency table. In order to monitor the multi-way contingency table data, this project tries to work on the following three fields: 1) Developing some new monitoring methods for sparse contingency table by applying Group Lasso method into a log-linear model and proposing some new monitoring statistics; 2) Proposing some new multivariate ordinal categorical control charts for the multi-way contingency table data with ordinal categorical variables by incorporating the ordinal information into an extended log-linear model ; 3) Developing some new approaches for monitoring multivariate autocorrelated contingency table data by applying a high-order Markov chain or context-tree method into a log-linear model.