Abstract:
|
At Amazon's Inventory Planning and Control Laboratory (IPC Lab) we run randomized controlled trials (RCTs) that evaluate the efficacy of in-production supply chain policies on important business metrics. Our customers are leading supply chain researchers and business managers within Amazon, and our mission is to help them best answer the question, 'Should I roll out my policy?' In this talk we discuss how we navigate multiple obstacles to fulfilling our mission. Our self-service system allows our customers to run hundreds experiments annually using data from our custom data store. Our experiments can contain hundreds of millions of observations, necessitating scalable statistical methodologies. Despite our large datasets, determining the correct data-driven decision is often difficult, as our data often exhibit high proportions of zeros, extreme right-tailed skewness, and small beneficial treatment effects. Finally, we discuss issues ranging from communication and translating experimental results to assessment of the financial impact.
|