Data Analytics for Supply Chain Sustainability

Learn how to use data analytics to manage a sustainable supply chain with practical tips and real-world examples.

(Image by Author)

Sustainable development is no longer an option. It is a necessity!

Companies must find ways to reduce their environmental impact while remaining profitable.

Example of Costs Reduction Initiatives — (Image by Author)

In this medium blog, I have shared analytics tools used in my career to support companies in their cost reduction initiatives.

As a data scientist, can you use data analytics for supply chain sustainability?

In this article, we will focus on using these tools and methodologies to reduce the environmental impact of your end-to-end Supply Chain.

I. Why do we need Supply Chain Sustainability?
II. Using Data Analytics to Improve Supply Chain Sustainability
  1. Descriptive Analytics Solutions
  Tracking Scope 3 CO2 Emissions of your distribution network with Python
  2. What is Life Cycle Assessment?
  Evaluate the environmental impacts of a product over its entire life cycle
  3. Generative AI X Sustainability: "The Supply Chain Analyst" GPT
  Smart agents to monitor operations with advanced analytics capacibilities
III. Resolve Operational Issues with Diagnostic & Prescription
  1. Identifying Root Causes with Diagnostic Analytics
  What is the main source of the emissions of last year?
  2. Optimizing for Sustainability with Prescriptive Analytics
  Linear Programming with Python to design the cleanest production footprint
  3. How Will Data Science Accelerate the Circular Economy?
  Overcome the operational challenges in transitioning to a circular economy
  4. What is Greenwashing? How We Can Use Analytics to Detect It
  Data analytics to detect and prevent greenwashing 
IV. Conclusion
  1. The Importance of Sustainable Supply Chains
  Footprint reduction became a must in the daily life of a supply chain manager
  2. Leveraging Data Analytics for Sustainability
  4 types of Supply Chain Analytics to improve the sustainability of your ops
  3. Green Inventory Management with a Digital Twin
  How can you reduce the emissions by reducing the frequency of stores delivery?
  4. Evaluating Circular Economy with Data Analytics
  Estimate the impact of a rental model on the emissions of a fashion retailer
  5. Reducing Environmental Footprint of Warehouse Operations
  Reduce waste and consumables usage in a warehouse

Introduction

The demand for transparency from investors in sustainable development has grown, and 2030 became the first significant milestone.

An increased emphasis on the sustainability of organizations has been placed via the ESG score.

Example of an international Supply Chain Network of a Fashion Retailer — (Image by Author)

In a previous publication, I presented Supply Chain Analytics as a set of tools to improve operations performance and reduce costs for an international fashion retailer.

Let us focus on reducing the environmental footprint of producing and delivering products to stores.

• How can we use descriptive analytics to measure the environmental impact of your supply chain?
• Can we optimize logistics flows to minimize CO2 emissions?
• How do you support your sustainable transformation with Generative AI?

We will answer in the following sections.

Using Data Analytics to Improve Supply Chain Sustainability

Supply Chain Analytics is a set of tools and methodologies using these information flows to answer questions and support decision-making.

Four Types of Supply Chain Analytics for Sustainability — (Image by Author)

Each type can answer specific questions and support your operations' green transition using methodologies and analytics tools.

What are the total CO2 emissions of your distribution network?

Tracking CO2 Emissions with Descriptive Analytics

A set of tools to provide visibility and a single source of truth across the supply chain to track your shipments, detect incidents and measure the performance of your operations.

❓ QUESTION
How many tons of CO2 are emitted by my distribution network?

This is the first step in your green transformation: you need to measure your operations' environmental impact.

A set of tools to provide visibility and a single source of truth across the supply chain to measure CO2 emissions:

By Country

Example of Report of CO2 emissions by country using PowerBI— (Image by Author)

By item

Example of Report of CO2 emissions by Item — (Image by Author)

The idea is to measure the CO2 emissions along the value chain and build reports to understand the impact of each leg of your distribution network.

💡 INSIGHTS
It looks like the easiest type from a mathematical point of view.
However, it may be the main bottleneck of your green transformation as it requires harmonizing data from different systems.

This is necessary to build a strong foundation for creating reports and feeding your advanced models for predictive, diagnostic, or prescriptive analytics.

Example of workflow of data processing for CO2 emissions reporting — (Image by Author)

📈 ANALYTICS
- Data extraction, transformation and processing with SQL, Python
- Visualizations using matplotlib and PowerBI
🚀 RESULTS
The measure of the emissions by scope and market to build your baseline of your transformation roadmap.

For more details, you can have a look at this article

Supply Chain Sustainability Reporting with Python

How much water is used to produce your T-shirt?

What is a Life Cycle Assessment?

As a second step, we can improve the reporting by

• Including emissions of the upstream flows up to the raw materials extraction
• Adding natural resource usage, waste generation and other pollutant emissions

Life cycle assessment (LCA) evaluates the environmental impacts of a product or service over its entire life cycle, from raw material extraction to disposal.

Resources Usage and CO2 Emissions along the Value Chain — (Image by Author)

This includes identifying the product being studied, the environmental impacts of interest, and the functional unit (the unit of measurement used to compare different products or services).

What is the environmental impact of producing and selling a t-shirt in a fast-fashion retail company?

Extract and process data from systems to feed your LCA report — (Image by Author)

From the analytics point of view, business intelligence and advanced data processing methodologies are used to collect and harmonize data.

For more details, have a look at this article

What is a Life Cycle Assessment? LCA

Have you heard about Generative AI?

Generative AI X Sustainability: “The Supply Chain Analyst” GPT

In another article, I share my explorative journey of Large Language Models used to improve the user experience of analytics models.

Supply Chain Control Tower Agent with LangChain SQL Agent [Article Link] — (Image by Author)

User: What is the percentage of CO2 emissions due to last-mile delivery?
Agent: 43% of the emissions (42 kTons CO2eq)

After OpenAI introduced a new feature allowing users to create custom versions of ChatGPT tailored for specific purposes, I took the opportunity to develop and deploy an agent to automate Pareto and ABC analyses.

“The Supply Chain Analyst” — (Image by Author)

The Supply Chain Analyst” is a custom GPT agent designed to automate supply chain analytics tasks and interact with users using natural language.

Could you imagine replacing a static “boring” dashboard with a prompt?

This agent can be linked to all the solutions presented in this article to optimize a user experience boosted by LLMs.

Create GPTs to Automate Supply Chain Analytics

Imagine all the possibilities!

Now that we have created solutions to visualize the past, let’s impact the future!

🏫 Discover 70+ case studies using data analytics for supply chain sustainability🌳and business optimization 🏪 in this: Cheat Sheet

Resolve Operational Issues with Diagnostic & Prescription

Identifying Root Causes with Diagnostic Analytics

This can be summarized as an incident or issue root cause analysis.

❓ QUESTION
What is the most polluting mode of transportation in France?

This step is very close to descriptive analytics. After building a single source of harmonized data, we can track the emissions along the value chain.

Distribution Emissions of transportation for two examples of store delivery — (Image by Author)

💡 INSIGHTS
If your colleague from the sustainability team is asking you why emissions doubled last year.
1. Have a look at the historical data 
2. Split the emissions by transportation mode (road, air, sea)

This will help you understand that transportation teams replaced sea freight with air freight after facing production delays.

🚀 RESULTS
This kind of diagnostic can help your management to spot the biggest sources of emissions and put its focus on them.
📊 ADDITIONAL KPIS
You can implement additional KPIs such as % of air freight (high emissions), % of electric last mile delivery to bring a more proactive approach and trigger alerts if needed.

What about Supply Chain Optimization?

Optimizing for Sustainability with Prescriptive Analytics

Assist the operations in solving problems and optimising resources to achieve CO2 emissions targets.

❓ QUESTION
Where should we locate our factories to minimize the CO2 emissions of our Supply Chain Network?

In the previous article, I presented solutions that assist operations in solving problems and optimizing resources to achieve the highest efficiency.

Supply Chain Network Design Article (Link) — (Image by Author)

For your green transformation, the tools and methodologies will be similar.

What is the optimal set of factories and warehouses to meet market demand at the lowest cost … OR … CO2 emissions?

The only difference is that your objective function will support the reduction of CO2 emissions.

Boolean Plot to Show the Results of the Supply Chain Network Optimization— (Image by Author)

I used methodology to design an optimal network of factories to manufacture and deliver products to specific markets.

Initial Linear Programming Problem — (Image by Author)

The idea was to use a linear programming model to select decision variables (factory locations) and produce and deliver products at a minimum cost.

💡 INSIGHTS
In this kind of exercises usually markets are very far from cheap production locations. Therefore, you'll have high levels of CO2 emissions due to transportation.

What if now we change the objective function?

New Problem Statement of Supply Chain Network Design Considering Sustainability — (Image by Author)

Let us now ask the model to minimize the CO2 emissions considering transportation and production while respecting the constraints of markets’ minimum supply.

Example of Sankey Chart representing the flows between factories and markets — (Image by Author)

🚀 RESULTS
You may need to completely transform your network to get plants close to your markets.
The impact on the cost will be not negligible as you'll have to produce in countries with high labor costs.
💡 GO BEYOND
Improve your model by including several types of transportation mode (electric, low emissions cargo fuels) that may increase the costs but reduce your CO2 footprint.
Thus, you may find alternative solutions and keep some of your facilities in competitive countries.

For more details, have a look at these articles,

Robust Supply Chain Network with Monte Carlo Simulation

Have you heard about Circular economies?

How Will Data Science Accelerate the Circular Economy?

A circular economy is a system where waste is minimized and resources are continuously reused or recycled.

Benefits of a Circular Economy — (Image by Author)

As the current linear economic model reaches its limits, discussions around new circular business models become more and more prominent.

What is holding us back?

The main blocking points are

• The operational and business obstacles blocking the transition
• Alternative strategies to increase the use of recycled materials
• Rental models to reduce the environmental footprint

As the analytics manager of a retail company, how can I support this transition?

Analytics experts can leverage the data generated by systems to overcome these barriers by identifying opportunities.

For more details, have a look at this article

How Will Data Science Accelerate the Circular Economy?

You are actually trying to reduce your emissions.

It’s not the case for everybody!

What Greenwashing Is, and How We Can Use Analytics to Detect It

Greenwashing is the practice of making misleading claims about the environmental benefits of a product or a service to communicate a false image of sustainability.

Five sins of greenwashing — (Image by Author)

Embellishing or hiding falsehood becomes challenging as companies seek environmentally conscious consumers' attention.

As a analytics experts, how can we help the world fight greenwashing?

Advanced Analytics used to detect greenwashing in environmental reports — (Image by Author)

The objective is to use

• Publicly available data: financial and sustainability reports, footprint databases, social media
• Advanced analytics models, including NLP, forecasting or statistical models to detect fraud

For more details,

What is Greenwashing? And How to Use Analytics to Detect It?

Conclusion

As a supply chain data scientist, you support Supply Chain managers.

Their scope of responsibilities includes

1. Understand and minimize the risks
2. Optimize operations to reduce costs and CO2 emissions
3. Provide visibility for resource planning
4. Prepare for future scenarios

For descriptive analytics, nothing changes except that now you need to put more emphasis on data quality.

And for the rest, adapt your focus and include CO2 emissions reductions in your objective functions.

More details in this video,

How to simulate the impact of initiatives?

Before implementing initiatives, companies need to forecast the return on investment.

This is where digital twins can help.

Simulating Scenarios with a Digital Twin

A digital twin is a digital replica of a physical object or system.

A Supply Chain digital twin is a computer model representing various components and processes involved in the supply chain, such as warehouses, transportation networks, and production facilities.

Example of Digital Twin with Python — (Image by Author)

Simulate several scenarios of green initiatives.

Scenario 1: You want to set up a local warehouse to reduce the last-mile delivery distance

• What would be the impact on the service level?
• What would be the impact of warehousing costs (with more locations)?
• How much CO2 emissions reduction can we reach?

Scenario 2: You would like to stop air freight to reduce your CO2 emissions

• What would be the impact on the stores’ replenishment lead times?
• How far ahead do the distribution planners need to create replenishment orders?
• What would the impact be on the reduction of CO2 emissions?

Scenario 3: You want to set up additional factories to produce locally for all markets

• What would be the impact on the production costs?
• What would be the impact of transportation costs (with factories close to the warehouse locations)?
• How much CO2 emissions reduction can we reach?

For each scenario, you can manipulate the parameter linked to the initiative and see how much your overall performance will decrease.

Example of element in your digital twin — (Image by Author)

Then, you can adapt the other metric (warehouse capacity and locations, replenishment lead time, …) until you reach the initial target.

What is the objective?

This will show you the improvements you need to make in your supply chain to make it robust enough to adapt to these new green initiatives.

💡 For more details,

What is a Supply Chain Digital Twin?

Let me give you an example.

What is the impact of delivery frequency on the CO2 emissions?

A logistics manager assumes that reducing this frequency would reduce emissions.

You can help him to verify this hypothesis.

Green Inventory Management with a Digital Twin

Green inventory management can be defined as managing inventory in an environmentally sustainable way.

Green Inventory Management Case Study — (Image by Author)

For a distribution network, this can involve a set of processes and rules that aim to reduce the environmental impact of order transmission, preparation and delivery.

Store Replenishment Logistics Process — (Image by Author)

What would be the impact on CO2e emissions if we reduce the frequency of store replenishments?

Use data analytics to simulate the variation of store replenishment frequency and measure the impact on the overall environmental impact.

• Reducing the number of trips per week
• Increasing the usage of large trucks that have lower emissions per ton
• Maximize the filling rate of trucks to avoid waste

The model presented in this article measures these impacts for multiple scenarios of delivery frequencies.

For more details,

Data Science for Sustainability- Green Inventory Management

Another cool example related now to circular economies.

Evaluating Circular Economy with Data Analytics

A circular economy is an economic model that aims to minimize waste and maximize resource efficiency.

Use Data Analytics to Evaluate the Sustainability of your Circular Economy — (Image by Author)

It involves designing products and processes focusing on longevity, reuse, and recycling.

Rental Subscription Model of Circular Economy — (Image by Author)

Some companies have implemented a subscription model where customers pay a regular fee to access a product or service for a specific period.

What is the impact of the subscription period on CO2 emission?

Use data analytics to simulate the impacts of several circular subscription model scenarios on a fast fashion retailer's emissions reductions and water usage.

We should rent this dress for a maximum of 22 days.

That’s the kind of insights you can get using the model presented in the article linked below, 👇

Data Science for Sustainability — Simulate a Circular Economy

Let’s take a step back in the warehouse.

Reducing Environmental Footprint with Local Initiatives

Managing waste and cutting consumables usage are also good ways to reduce the warehouse operations' footprint for store deliveries.

Reduce Wrapping Film Usage — (Image by Author)

Data analytics can also help you estimate the impact of these small initiatives on the overall network (if applied to all warehouses).

1. Calculate the current consumption using actual operational data
2. Include these parameters in your digital twin
3. Design a Proof Of Concept to estimate the savings target
4. Extrapolate the savings (per warehouse) to the whole network

All these initiatives align with the efforts of cost reduction and operational excellence.

Can we estimate the impact of these initiatives with data analytics? Yes!

You can adapt your continuous improvement methodologies to track consumables usage and find alternatives that will not impact your productivity.

First, start with measuring your operations' current consumption and environmental impact.

Data Analytics can help you to automate this process, more details in the article linked below 👇

Sustainable Logistics - Reduce Warehouse Consumables

About Me

Let’s connect on Linkedin and Twitter. I am a Supply Chain Engineer who uses data analytics to improve logistics operations and reduce costs.

For consulting or advice on analytics and sustainable supply chain transformation, feel free to contact me via Logigreen Consulting.

If you are interested in Data Analytics and Supply Chain, look at my website.

Samir Saci | Data Science & Productivity

💌 New articles straight in your inbox for free: Newsletter
📘 Your complete guide for Supply Chain Analytics: Analytics Cheat Sheet