Information gathered by low-cost sensors can improve logistic decision-making. The eco-efficient supply chain project investigates how this information can be used to reduce waste and carbon emissions for better managing perishable goods.

Abstract

Loss rates of perishable goods are significant. Statistics by the U.S. Department of Agriculture report that average yearly loss rates at the retail level alone can be as high as 51% for specific commodities. When reviewing the loss figures from a quality management perspective, it is hard to think of any industry that could survive with such high defect rates.

The limited shelf-life of perishable goods and their susceptibility to fluctuations in environmental parameters often force managers to make decisions under uncertainty, leading to high loss rates. In order to achieve the same desired output, additional input resources are often required for compensation.

Low-cost measurement technologies such as Radio Frequency Identification tags and sensors could provide the necessary information to make the quality deterioration process more predictable and to make problems visible.

This project focuses upon gaining more knowledge of the value of sensor information in this context and aims to lay the groundwork for future research. The main hypothesis postulates that a significant part of compensatory additional input resources can be substituted with better information.

Working group

Alexander Ilic, Thorsten Staake, Elgar Fleisch

Description

Early results of this research project show that sensor information indeed helps increasing efficiency in retail operations and sourcing planning.

In retail operations, active removal and sorting decisions at the store level can lead to significant waste reduction and an increase of perceived quality. Moreover, if these sensor-based decisions are moved upstream the supply chain, significant reductions of greenhouse gas emissions can materialize due to better load utilization.

When looking at sourcing decisions, sensor information provides means to reveal hidden costs associated with route-specific loss rates. This ultimately improves sourcing decisions by disclosing and thus preventing otherwise hidden problems. The result is a maximization of resource efficiency, clear incentives for improved cool chain handling, and a reduction of waste.

As a key contribution to theory, these results confirm that sensor information can improve the management of perishable goods and can replace additional input resources with better information. This result underlines the relevance of integrating item-level quality modeling into the research stream of operations research. It also opens up new research fields on questions such as the accountability of losses along supply chains and their implications for contracts and buy-back options.

For practitioners, this project provides a framework to review their operations and planning processes, to establish new key performance indicators, and to identify critical product groups for an evaluation with sensor-based monitoring. Moreover, this project highlights the relevance for retail supply chains to rethink their data management strategies and to consider environmental externalities.

For governments and public organizations, this project provides recommendations on increasing their data intelligence. Federal organizations such as the USDA should switch from collecting yearly average loss rates to geospatial figures with monthly resolution. This change would also incentivize agricultural belts to improve their performance. As a consequence, federal funds for building up economic areas can be distributed in a more effective way, leading to increased food system security and potentially a rebirth of regional food systems.

Contact

If you are interested in Eco-efficient Supply Chains, please contact Thorsten Staake (tstaake@ethz.ch).