Home
The aim of the BONSAI network is to make reliable, unbiased sustainability information on products – “product footprints” – readily and freely available whenever and wherever it is needed to support product comparisons and decisions.
BONSAI is a not-for-profit organization, with the objective to maintain open access data and open source software and algorithms to produce “product footprints”.
Thinking of contributing? Head to the getting started readme for guidance.
Page contents
The BONSAI Open Database
Bonsai Wiki description
Wiki contents overview
Main distinguishing features of BONSAI
The BONSAI project aims to build a shared resource where the community can contribute to data generation, validation, and management decisions. In particular, its main goal is to maintain an Open Database of LCA data and an open source toolchain capable of supporting LCA calculations. This will allow the science of lifecycle assessment to perform in a more transparent and more reproducible way, and will foster data integration and sharing. Supporting in this way the use of LCA studies in democratic decision making.
Linked Open Data and semantic technologies are a natural choice for achieving this goal. The BONSAI Database is a community resource built around (1) a comprehensive ontology for industrial ecology and associated relevant data; and (2) the Linked Open Data database and endpoint aggregating and distributing the collected data.
Currently we have developed an official ontology to model all the core aspects related to the domain of Life Cycle Assessment. Moreover, we are in the process of converting many distinct datasets (e.g., Multi-Regional Environmentally Extended Supply-Use Tables, as well trade and supply chain inventory data. The current converted data is available as linked open data on the official SPARQL endpoint.
The BONSAI wiki is the result of a collective ongoing effort by the developers community. It contains a draft description of the functional and technical specifications of the project, as described below.
-
Functional specifications: a description of the specific functions that a component must perform. The purposes is to achieve a team consensus on what the program is to achieve before moving to more time-consuming effort of writing source code.
-
Technical specifications: a description of the technical requirements to fulfil the function defined by the functional specification, e.g.: software to be used, inputs that can be provided to the software system and how the system responds to those inputs. The purpose is to facilitate the harmonisation of different components and avoid errors due to lack of compatibility and interoperability issues.
The pictures below draft the BONSAI flow diagram according to potential working groups. Additional descriptions are provided for each Working Groups in the following wiki-pages.
Simplified structure of the BONSAI database. See Glossary for definitions
Visit BONSAI website for a more exhaustive description of the BONSAI vision, organisation and strategy.
The following pages are dedicated to the fundamental tasks for the completion of the BONSAI Beta version, a future complete version to be used for testing purposes:
Some of the Core Database Architecture tasks correspond to Working Groups and the respective Work Packages. In this case, the task description is in the Work Package. The Working Groups are distributed over 3 action areas:
Harvest data work packages:
Input Output Framework
Big Data harvesting
Data Provider Interface
Ensure data quality work packages:
Global Impact Assessment
Uncertainty
Validation
Data review
Make data usable work packages:
Product System Algorithms
Data Access
User Communities
Compared to existing databases and software to produce product footprints, the BONSAI project is distinguished by:
- Being based on open data and open code;
- Having a low barrier to data entry supported by an open data review process;
- Completeness of data, obtained by integrating Multiregional Input Output (MRIO) and detailed process-based data;
- Flexible addition and editing at the level of single data triples;
- The natural language query interface for final product footprints;
- Transparency of the underlying data and routines;
- The modular approach to software design.