In this research study the researcher proposed a predictive model for Automation Performance on Logistics Companies Performance with Perishable Goods. In this context pick and pack including delivery at doorstep is one the challenging issues in the real words. The automation technologies have a potential role to automate the business process and provide secure and transparent way perishable goods and services. In current business industries the companies looking for fact and secure automation system to operate the business process in logistics industries. To identify the automation performance of logistics system the researcher used the 395000 datasets with different supervised machine learning algorithms to automate the logistics process During the research study the researcher found that Logistics Classifiers training accuracy 87.97% and testing accuracy 87.62%, Random Forest Classifiers training accuracy 98.59% and testing accuracy 9.29%, Support Vector Machine (SVM) training accuracy 97.48% and testing accuracy 97.85%, KNN Classifiers training accuracy 53.37%and testing accuracy 43.51%, Decision Tree classifiers training accuracy 98.68%and testing accuracy 98.36%, ADA Booster training accuracy 74.3% and testing accuracy 75.02%, XGB Booster training accuracy98.62%and testing accuracy 98.90%.Finally the researcher concluded that decision tree classifier and XGB Booster predictive model are having the 98% accuracy level and that predictive model would be suitable to measure the automation performance on logistics companies.