# 57: Top 3 Applications of Data Science to Transform Warehouse Operations

May 15, 2023

Hello,

I wrapped up the Spring Semester I taught at NJIT’s business school this week. It was a lot of work (content creation, assignments, and student interaction) over the past 15 weeks, but I loved it! (well, except for fielding queries on grading😫)

While I’ve delivered short guest lectures at several universities, this experiment was to check if I’d enjoy a more intensive teaching experience. Helping students demystify data analytics, understand industry applications, and learn experientially was fulfilling.

I’m extending this experiment and accepting the invitation to teach the same course over the upcoming summer semester. I’ll share my learnings in the coming weeks.

This newsletter will take you about 4 minutes to read.

I. Spotlight: Top 3 Applications of Data Science to Transform Warehouse Operations

In traditional warehouses, mechanical issues, fluctuating weather, pick-up delays, and traffic can make accurate scheduling and staff planning a complex guessing game, causing long turn times and increased facility costs due to overstaffing and fines.

Covid-19 has exposed the inefficiency of supply chains and the importance of building their resilience for the future. Now, the demand driven by e-commerce retailers means that traditional warehouses are being replaced by next-generation warehouses that can deliver predictive, cost-effective, and innovative data-driven services.

Tracking transportation metrics and data patterns has not been exploited to its full potential, for example, to better predict incoming truck arrivals. But truck turnaround times can be improved and supply chains streamlined by predicting carrier appointments using Artificial Intelligence (AI) and machine learning (ML).

Shippers, logistics, and supply chain leaders looking at data-driven applications to help modernize and automate warehouse operations while improving customer satisfaction should consider these three powerful data science applications.

a large stack of red, blue, and orange bins — Photo by Yuriy Vertikov on Unsplash

Demand and Supply Optimization

To effectively plan the staffing capacity on any given day, warehouses must forecast the demand of pickups to assign enough warehouse pickers to carry out the operational activities.

As staffing capacity is based on the number of hours per picker and the number of pickers, supervisors traditionally tweak these levers manually based on prior experience. Often, pickups are impacted due to fluctuations in warehouse demand and variability in staffing capacity. This disrupts customer operations

A data-driven solution forecasts the demand ahead of time, and plans staffing capacity with a high degree of…

II. Industry Roundup:

1. Article: The Next Generation Of Large Language Models

15 minutes | Forbes | Rob Toews

The author says that the frontiers of AI are advancing rapidly, and ChatGPT is a mere stepping stone to what comes next. This article highlights three emerging areas that will define the next wave of innovation in Generative AI and LLMs (large language models). These are 1) models being able to generate their own training data to self-improve, 2) models that can fact-check themselves, and 3) the widespread prevalence of sparse expert architectures.

-> Read the Article

2. Interview: "Godfather of artificial intelligence" talks potential and risks of AI

43 minutes | YouTube | CBS Mornings

In this insightful interview, AI pioneer, Geoffrey Hinton, demystifies how deep learning works. He says there is still a divergence between AI and the human brain’s functioning. He discusses the ethical dilemmas of making autonomous lethal weapons. He says computers will soon come up with original ideas for improving themselves and why that could spell danger for humanity. (Update: Earlier this month, Hinton quit Google, saying he regrets his contribution to the field)

-> Watch the video

III. From my Desk:

1. Podcast: What’s the hype vs Reality in Generative AI?

1 min | Pondering AI podcast

Should we be so excited about (Generative) AI today? What's the hype vs reality, and how do we draw the line? I spoke to Kimberly Nevala for the Pondering AI podcast. We pondered over business expectations while implementing AI, how Decision Intelligence can transform org decisions, and where Data Storytelling plays a crucial role in all of this. Give it a listen, and share what resonated the most with you!

-> Listen to the Podcast

2. Hackathon: Health AI Finals at Northwestern University

1 min | LinkedIn

Last week, I was part of the judge panel in the AI competition finals organized by the Institute for Augmented Intelligence in Medicine. Top 5 teams from various universities pitched their solutions in the 3rd AI Health Bowl. Built over six months, the teams addressed some of the big healthcare disparities in the world using data & analytics.

-> Check out the Winning Ideas

This robot has realistic facial expressions & talks about feeling alive - what do you think? 🤔

Thank you for subscribing and reading the newsletter. I appreciate your attention,

Ganes.

PS: Did someone forward this to you? You can subscribe here.

My Website | LinkedIn | Twitter | YouTube

I’m Ganes Kesari. I publish ‘Data-Driven Future’ to help understand how data shapes our world, explore key trends, and explain what they mean for you today. I speak and write to demystify data science for decision-makers and organizations.

Recent Issues:

Our Data-driven Future

Discussion about this post