Premium
ODYM—An open software framework for studying dynamic material systems: Principles, implementation, and data structures
Author(s) -
Pauliuk Stefan,
Heeren Niko
Publication year - 2020
Publication title -
journal of industrial ecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.377
H-Index - 102
eISSN - 1530-9290
pISSN - 1088-1980
DOI - 10.1111/jiec.12952
Subject(s) - industrial ecology , material flow analysis , computer science , modular design , python (programming language) , material flow , software , data structure , workflow , software engineering , object oriented programming , systems engineering , industrial engineering , distributed computing , database , engineering , programming language , ecology , sustainability , biology , waste management
Abstract Material flow analysis (MFA) studies the stocks and flows of goods and substances in systems. The methods and algorithms of MFA have improved over the last few years, but a flexible platform that integrates recent modeling advances such as simultaneous consideration of the product, component, material and chemical element levels, lifetime models, and uncertainty treatment is not available. There is also no versatile data format for exchanging data between projects. This lack of research infrastructure is detrimental to scientific progress. To fill that gap, we propose a novel industrial ecology community model for MFA. The Open Dynamic Material Systems Model (ODYM) is an open source framework for material systems modeling programmed in Python. The description of systems, processes, stocks, flows, and parameters is object‐based, which facilitates the development of modular software and testing routines for individual model blocks. ODYM MFA models can handle any depth of flow and stock specification: products, components, sub‐components, materials, alloys, waste, and chemical elements can be traced simultaneously. ODYM features a new data structure for material flow analysis; all input and output data are stored in a standardized file format and can thus be exchanged across projects. It also comes with an extended library for dynamic stock modeling. We present the features, design principles, software, and data structure of ODYM, describe its main methods and functions, and give an outlook on current and future applications.