Variability of temperature exposure of dairy products during refrigerated e-commerce distribution and its impact on dairy quality.

Abstract

E-commerce food distribution has grown drastically in recent years, a trend that was accelerated by the COVID-19 pandemic. The range of perishable products purchased through e-commerce as well as the distance over which these products travel to consumers has also increased considerably. Perishable goods are commonly available through various e-commerce channels (e.g., third-party grocery shopping and delivery, direct processor-to-consumer delivery, and overnight shipment via centralized distribution centers). These e-commerce distribution methods have grown to accommodate the increasing demand for grocery delivery, which can introduce the potential for temperature abuse of these perishable goods, possibly leading to premature spoilage and quality deterioration. To determine the effect of e-commerce distribution on the shelf life of dairy products, we assessed the variability of simulated and real e-commerce time-temperature profiles of fluid milk and Greek yogurt in 3 transportation channels: (1) direct-to-consumer, (2) distributor or business-to-consumer, and (3) business-to-business-to-consumer. To further identify how real dairy products ordered through business-to-business-to-consumer (e.g., grocery delivery to consumers from retail stores through a third party) channels, containers of 1.89 L (1/2 gal) milk and 157 mL (5.3 oz) Greek yogurt were delivered to customers from local retail chains, and temperatures were measured upon delivery. Finally, the temperature profiles measured during 1.89 L milk transportation were used to conduct microbial and sensory shelf-life testing on commercial containers of milk and were also incorporated into a previously developed computational model that predicts milk spoilage based on the initial concentration of primary groups of bacterial contaminants and other relevant conditions (e.g., storage temperature). Our time-temperature profile results showed that product temperature at the time of delivery ranged from 0.2 to 10.1°C for the direct-to-consumer pathway, -0.9 to 19.2°C for the business-to-consumer pathway, and 3.1 to 18.3°C during the simulated business-to-business-to-consumer pathway. Average milk and yogurt arrival temperatures from real business-to-business-to-consumer deliveries were 7.2°C and 7.3°C, respectively, and ranged from 3.1 to 10.5°C and 3.4 to 12.7°C, respectively. Results from fluid milk shelf-life microbial testing and end of shelf-life sensory testing showed limited effect following simulated short-term temperature exposure from e-commerce conditions, although more data are needed. This is supported by our model, which also indicates that there are minimal changes to expected microbial spoilage of fluid milk at the end of shelf life. However, our models only predict microbial growth and not sensory quality, which has a larger effect on consumer acceptance and should be assessed in future studies. Overall, our study provides important information on the exposure of dairy products to temperature variability during e-commerce distribution, which can be used to further develop strategies for controlling and monitoring the cold chain.