Abstract
Motivation
Generalized Read-Across (GenRA) is a data-driven approach to estimate physico-chemical, biological or eco-toxicological properties of chemicals by inference from analogues. GenRA attempts to mimic a human expert’s manual read-across reasoning for filling data gaps about new chemicals from known chemicals with an interpretable and automated approach based on nearest-neighbors. A key objective of GenRA is to systematically explore different choices of input data selection and neighborhood definition to objectively evaluate predictive performance of automated read-across estimates of chemical properties.
Results
We have implemented genra-py as a python package that can be freely used for chemical safety analysis and risk assessment applications. Automated read-across prediction in genra-py conforms to the scikit-learn machine learning library's estimator design pattern, making it easy to use and integrate in computational pipelines. We demonstrate the data-driven application of genra-py to address two key human health risk assessment problems namely: hazard identification and point of departure estimation.
Availability and implementation
The package is available from github.com/i-shah/genra-py.
PurposeDistribution of long-haul goods could be managed via multimodal transportation networks where decision-maker has to consider these factors including the uncertainty of transportation time and cost, the timetable limitation of selected modes and the storage cost incurred in advance or delay arriving of the goods. Considering the above factors comprehensively, this paper establishes a multimodal multi-objective route optimization model which aims to minimize total transportation duration and cost. This study could be used as a reference for decision-maker to transportation plans.Design/methodology/approachMonte Carlo (MC) simulation is introduced to deal with transportation uncertainty and the NSGA-II algorithm with an external archival elite retention strategy is designed. An efficient transformation method based on data-drive to overcome the high time-consuming problem brought by MC simulation. Other contribution of this study is developed a scheme risk assessment method for the non-absolutely optimal Pareto frontier solution set obtained by the NSGA-II algorithm.FindingsNumerical examples verify the effectiveness of the proposed algorithm as it is able to find a high-quality solution and the risk assessment method proposed in this paper can provide support for the route decision.Originality/valueThe impact of timetable on transportation duration is analyzed and making a detailed description in the mathematical model. The uncertain transportation duration and cost are represented by random number that obeys a certain distribution and designed NSGA-II with MC simulation to solve the proposed problem. The data-driven strategy is adopted to reduce the computational time caused by the combination of evolutionary algorithm and MC simulation. The elite retention strategy with external archiving is created to improve the quality of solutions. A risk assessment approach is proposed for the solution scheme and in the numerical simulation experiment.
Generalized Read-Across prediction using genra-py
