Mapping and valorizing food loss and waste data in the Amsterdam Metropolitan Area to improve the circular economy (AMAFLOW)

  • Home
  • Mapping and valorizing food loss and waste data in the Amsterdam Metropolitan Area to improve the circular economy (AMAFLOW)

Mapping and valorizing food loss and waste data in the Amsterdam Metropolitan Area (AMA) to improve the circular economy (AMAFLOW)

Key Objective

Improve the food loss & waste monitoring system in AMA and valorise the registered waste data to create values for data economy

Challenge Addressed

  • Identify the food waste flows from the general waste flows registered in Landelijk Meldpunt Afvalstoffen (LMA) data to improve the food losses & waste monitor in AMA.
  • Valorise the improved FLW data to develop a spatial model that can optimize the reverse logistic networks for the high potential waste flows in AMA
  • Valorise the improved FLW data to create a biodiversity footprint calculator for AMA with respect to FLW.

Type of Stakeholders

Amsterdam municipality, Amsterdam institute for advanced metropolitan solutions, waste and food companies in AMA
Case Study Acronym:AMAFLOW
Long Title of the Case Study:Mapping and valorizing food loss and waste data in the Amsterdam Metropolitan Area (AMA) to improve the circular economy
Case Study Main Contact:

Xuezhen Guo

Countries involved and main place of the Case Study:Amsterdam Metropolitan Areas
Part of the Food System addressed:Losses and wastes along the food supply chains in AMA

Case Study Summary

Automatically identify and extract the food components from the LMA dataset in as much as possible details to develop a more reliable food loss & waste (FLW) monitor to provide the basis for further data valorization.

APIs will be created combined with the text mining and automatic metrics calculation algorithms to generate the food loss & waste dashboards and maps for AMA. This will ensure we always have the up-to-date data.

Required Data and procedure to estimate the waste composition

In the ideal situation, we want to have the description of each individual waste flow registered in the LMA database. If this is possible, then we can use text mining to identify the meaningful texts in the large amount of records and categorize them into e.g., different food items.

If the low-level data is not available, we will at least have the names (or even description) of the companies. Since the SBI codes do not cover all the registered companies in the databases (the aggregate data of 2019 show that a large number of flows cannot be categorized using the SBI code). Then we can use text mining to classify the flows (e.g., matching the companies with their web-information). Domain expert knowledge will also be used to improve the accuracy of the machine learning algorithms.

Biodiversity footprint of the FLW in AMA 

After having the more reliable food waste database, then we can use it to calculate many other metrics. We will start with calculating the biodiversity footprint of FLW in AMA. We use the local food production and food import data to create a “food sourcing profile” for AMA. Then combining the land use data of the “source country” for each food item with the “food sourcing profile”, we can calculate the biodiversity footprint of the FLW in AMA using the model that specifies the relationship between the land use and biodiversity losses. 

Optimize the reverse logistic network

A spatial model combining the regression and optimization techniques will be developed to optimize the reverse logistic networks for the identified high potential waste streams regarding e.g., nutrient values, biodiversity impacts. The model will deliver the optimal locations of the waste collection centers and treatment plants as well as the most suitable operational schedules for waste transportation and processing. 

Motivation and key expected Outcome

Food Loss and Waste (FLW) is a big issue related to food security, climate change, biodiversity, etc. It has been prioritized in UN sustainable development goals (SDG) target 12.3 to contribute to “ensure sustainable consumption and production patterns”. Since cities are densely populated areas, urban food waste is a hotspot for food waste monitoring and valorisation.

The key expected outcome of this research include an improved FLW monitor in AMA, a biodiversity footprint calculation tool for FLW, as well as a spatial optimization model to design the reverse logistic networks for high-potential FLW streams.

Main Partners of the Case Study

Wageningen Food & Biobased research (WFBR):
Project lead, food waste management expert, modelling experts with optimization techniques and machine learning.

Xuezhen Guo 

Other Stakeholders involved in the Case Study realisation, but not direct project partners

geoFluxus: The data expert which manages the waste databases from LMA and will work together with WFBR to identify FLW flows, improve the data quality and facility modelling with data inputs. 

Arnout Sabbe   

Amsterdam institute for advanced metropolitan solutions: Link to stakeholder in AMA including policy makers, companies, etc. 

Willie van den Broek

Amsterdam Municipality: facilitate the development of the monitor and tools by providing policy relevant insights

Addressing the
Food Value Chain

The food value chain considered in this research covers the stages of “primary production”, “storage”, “transportation”, “processing”, “distribution” and “consumption”. It deals with the food wastes generated in those chain stages and the strategies to recycle and reuse them in AMA.

Cross collaboration
with other Projects or Initiatives

We have several on going projects that are relevant to this project. In WFBR, we have 4 ongoing projects from the ministry of agriculture about developing the national monitor for FLW in the Netherlands. We also have many “kennisbasis” and “BO” projects addressing food circularity and biodiversity calculations.

Your submission was successful.

Thank you!