ups-leverages-bigquery-and-striim-for-ai-secured-package-delivery

UPS leverages BigQuery and Striim for AI-secured package delivery

UPS Capital® is leveraging Google’s Data Cloud and AI technologies to safeguard packages from porch piracy. With more than 300 million American consumers turning to online shopping, UPS Capital has seen its customers face significant challenges in securing their package delivery ecosystem. Now, the company is leveraging its digital capabilities and access to data to help customers rethink traditional approaches to combat shipping loss and deliver better customer experiences.

Safeguarding shipments with AI and real-time data

Established in the 1990s, UPS Capital has undergone a significant transformation from a financial services division to a multifaceted suite of solutions dedicated to protecting businesses’ goods in transit. With a range of offerings, including innovative insurance, bolstered carrier protection, and integration support for top ecommerce platforms, UPS Capital empowers businesses to navigate shipping with assurance and ease while reducing risk and preserving cash flow.  

The exponential rise in package deliveries, combined with the increased value of ordered items, presents an attractive opportunity for thieves and has led to a spike in shipping loss and insurance claims. A 2022 TransUnion report indicated shipping fraud surged over 1,500% from 2019 to 2021 alone. Traditionally, UPS Capital has offered customers insurance products that help serve as a financial backstop against shipping loss. With the advancement of computing capabilities and AI, UPS Capital has been able to take a new approach to risk mitigation with the predictive intelligence of DeliveryDefense™ Address Confidence.

Video Thumbnail

Leveraging Google’s Data Cloud and AI technologies, DeliveryDefense Address Confidence utilizes real-time data and machine learning algorithms to safeguard packages. By assigning a confidence score to potential delivery locations, it enhances the assessment of successful delivery probabilities while mitigating loss or theft risks. Every address is allocated a confidence score on a scale from 100 to 1000, with 1000 indicating the highest probability of delivery success. These scores are based on customer reports of package theft. Shippers can integrate this score into their shipping workflow through an API to take proactive, preventative actions on low-confidence addresses. 

For instance, if a package is destined for an address with a low confidence score, the merchant can proactively reroute the shipment to a secure UPS Access Point location. These locations typically have a confidence score of around 950 due to their high chain of custody security precautions.

“At UPS, we’re leveraging BigQuery, Looker and Vertex AI to create products that protect customers from risk by helping them avoid it in the first place. Integrating DeliveryDefense Address Confidence scores into the shipping process helps customers save money through reduced losses, save time with a reduction in the amount of manual order reviews, and improve the customer experience with safe and convenient shipping options.” – Pinaki Mitra,  Vice President, Data Science & Machine Learning, UPS Capital 

One of UPS Capital’s customers, SunJoy, has reduced its losses by 35% with DeliveryDefense by redirecting shipments to a safer location or adding an adult signature in response to alerts about high-risk deliveries.

By streamlining operations and improving delivery fulfillment rates, this service also fosters customer satisfaction from timely and reliable deliveries, lower costs, and overall improved service quality. The ripple effect of this spreads across the entire supply chain, positively impacting all involved parties and contributing to a more resilient and competitive marketplace. 

The technical backbone of UPS DeliveryDefense

https://techontheblog.com/wp-content/uploads/2024/05/localimages/UPS__Striim_Blog_-_Architecture_Image.max-1100x1100.png

“The unified data cloud from Google Cloud simplifies the entire data lifecycle, making it easier for businesses to manage their data effectively. By seamlessly integrating different datasets into BigQuery and utilizing advanced tools like Vertex AI, organizations can unlock valuable insights without unnecessary complexity.” – John Kutay, Director of Product, Data & AI, Striim

UPS DeliveryDefense Address Confidence relies on a robust technical infrastructure that begins with the direct ingestion of diverse datasets into BigQuery. BigQuery serves as the cornerstone repository for structured data, providing a reliable data warehousing solution within Google Cloud. Simultaneously, data sourced from SQL Server undergoes meticulous cleansing. These harmonized datasets, enriched with multimedia elements, are combined in BigQuery and seamlessly integrated with other Google Cloud services. 

Next, Vertex AI takes the stage, empowering the execution of intricate machine learning models, such as route anomaly detection and fraud detection models for shipping transactions. Using Vertex AI’s vast array of tools, UPS Capital can train, refine, and deploy models to unearth insights, predict trends, and extract valuable knowledge from its data. Firestore, a flexible, scalable NoSQL database for mobile, web, and server development from Firebase and Google Cloud, catalogs insights, confidence scores, and analytical details — all made accessible through the Looker API.

Generative AI in real-time data streams with Striim and Google Cloud

“Striim and Google Cloud have jointly enabled us to enhance the customer experience with AI and ML.” – Pinaki Mitra,  Vice President, Data Science & Machine Learning, UPS Capital 

AI-driven innovation is underway at Striim, particularly in the domain of gen AI and streaming contexts. The real-time data integration platform compliments Google Cloud’s modern architecture by dynamically embedding vectors into streaming information, enhancing data representation, processing efficiency, and analytical accuracy. Striim also integrates structured and unstructured data pulled from diverse sources and applies a variety of AI models from OpenAI and Vertex AI to generate embeddings that establish similarity scores between data points to reveal possible relationships. 

Throughout this process, the streaming data is enriched with dynamically generated vectors, setting it apart from traditional batch processing methods. Unlike its predecessors, which rely on post-collection analysis, Striim’s innovative technique operates instantaneously, continuously updating and refining the data flow. By embedding vectors into the streaming process, the system not only enhances the relevance of responses to queries but also catalyzes ongoing model improvement through the influx of real-time data. 

UPS Capital selected Striim and Google Cloud to enable the use of AI models with real-time data streams. As a result, the company is able to continuously improve its defense models at the speed of business while reducing the latency of detecting suspicious activity. By analyzing streaming data and dynamically generating vectors, UPS Capital can enhance the accuracy and effectiveness of its defense mechanisms over time, ensuring ongoing protection against changing and evolving threats.

Future-Ready: UPS’s defense strategy in package delivery

UPS Capital’s DeliveryDefense Address Confidence, powered by Google BigQuery and Striim, transforms package delivery security through real-time data and machine learning. By assigning confidence scores and leveraging proactive measures, UPS Capital helps shippers mitigate risk and improve customer experience. Discover how Striim on Google Cloud can empower real-time intelligence for AI, just like UPS’s DeliveryDefense Address Confidence. Sign up for a free trial today!

Posted in

Share this article
Shareable URL
Prev Post

Cloud SQL: Rapid prototyping of AI-powered apps with Vertex AI

Next Post

Game-changing assets: Making concept art with Google Cloud’s generative AI

Leave a Reply

Your email address will not be published. Required fields are marked *

Read next