• Gain aggregated insights from the acquired data through advanced analytics, such as pattern recognition and predictive analytics 
• Understand the root-cause of deviations faster through data-driven scenario analysis 
• Improve operational efficiency of the value chain while managing risk by applying predictive models 
• Support and automate the decision-making process based on trusted data through system generated suggestions

Active or passive temperature management systems are commonly used when shipping sensitive goods. However, such temperature management systems come with drawbacks in terms of operative handling and cost compared to standard packing materials.

What if using standard packaging would be sufficient based on favorable external temperature during shipment?

In collaboration with a client, we sampled more than 800 shipment routes across all regions of Switzerland during winter, to assess the potential for a risk-based approach to use conventional packaging for the shipment of medicinal products. We wanted to know how many shipments would not require full thermal insulation, but only reduced insulation or even none at all (standard packaging).

The Results

The result was astonishing: taking a cold-chain shipment with a temperature band of 2 to 8 °C, we found that approximately 40% of these shipments had the potential for optimization.

To offer our client an operative optimization tool, we created a model to predict the package temperature during transit using machine learning and created a first demonstrator: the MODsight Packaging Selector.