To reduce sustainability, the need for long-term sustainable practices has never been more critical.
How do we use proxy AI-enhanced analytics to support companies in green transformation?
For years, my blog has focused on using supply chain analytics methods and tools to solve specific problems.
At Logigreen, the startup I founded, we deployed these analytical solutions to help retailers, manufacturers and logistics companies meet their sustainability goals.
In this article, I will demonstrate how to use AI proxy to enhance these existing solutions.
The goal is to make it easier and faster for companies to implement sustainability plans across supply chains.
Obstacles to the green transformation of a company
As political and financial pressures shift away from sustainability, green transformation is made easier and easier to obtain.
Last week, I participated in the world ChangeNow The meeting was held in my hometown of Paris.
The conference brings together innovators, entrepreneurs and decision makers who, despite the challenges of the background, are committed to building a better future.
This is a great opportunity to meet some of my readers and connect with leaders who drive change across the industry.
Through these discussions, a clear message emerged.
There are three major obstacles for companies to drive sustainable transformation:
- Lack of visibility into the operation process,
- The complexity of sustainability reporting requirements,
- Challenges of designing and implementing programs throughout the value chain.
In the following sections, I will explore how we use it Agent AI Overcome two of these major obstacles:
- Improve reports to respect regulations
- Accelerate the design and execution of sustainable plans
Solve AI Agents Reporting Challenges
The first step to any sustainable roadmap is to build a report foundation.
Companies must measure and publish their current environmental footprint before taking action.
For example, ESG report communication Environmental performance of the company (E)Social responsibility (S)and governance structure strength (g).
Let’s start by solving the data preparation problem.
Question 1: Data collection and processing
However, many companies face major challenges from the start, Data collection.
In the previous article, I introduced the concept of life cycle assessment (LCA) – a method to evaluate the environmental impact of a product from raw material extraction to disposal.
This requires complex data pipelines to connect multiple systems, extract raw data, process it and store it in a data warehouse.
These pipelines help generate reports and provide a unified data source for analytics and business teams.
How do we help non-technical teams browse through this complex landscape?
In Logigreen, we explore the usage of AI proxy for text-to-SQL applications.
The added value is that business and operations teams no longer rely on analytical experts to build tailored solutions.
As my own supply chain engineer, I understand the frustration of operations managers who have to create support tickets just to extract data or calculate new metrics.
With this AI agent, we provide an analytical service experience to all users, allowing them to make demands in simple English.
For example, we help the reporting team build specific tips to collect data from multiple tables to feed the report.
“Please generate a table to show the sum of the daily emissions from all delivery from warehouse xxx.”
For more information on how I implement this proxy, please check out this article.
Automatic supply chain analysis workflow using N8N | Towards Data Science
Issue 2: Report format
Even if the data is collected, companies face another challenge: Generate a report in the required format.
In Europe, new Company Sustainability Reporting Directive (CSRD) Provides a framework for companies to disclose their environmental, social and governance impacts.
According to CSRD, the company must XHTML format.
This document is full of detailed information ESG taxonomya process that can be highly technical and error-prone is required, especially for companies with low data maturity.
Therefore, we have tried to automatically review reports using AI agents and provide summary to non-technical users.
How does it work?
Users send reports via email.
The endpoint will automatically download attachments, review content and formats, and search for errors or missing values.
The results are then sent to an AI agent, which generates a clear audit summary in English.
Agent sends the report back to the sender.
We have developed a fully automated service for audit reports created by sustainability consultants (Our client is a consulting company) Anyone can use it without the need for technical skills.
Interested in implementing a similar solution?
I built this project using the codeless platform N8N.
You can My N8N creator profile.
Now that we have explored the reporting solutions, we can continue to develop the core of green transformation: Design and implement sustainable plans.
Agent AI for Supply Chain Analysis Products
Sustainability analysis products
Over the past two years, my focus has been on building analytics products, including web applications, APIs, and automated workflows.
What is the sustainability roadmap?
In my previous experience, it usually starts with the push of senior management.
For example, leadership will ask the supply chain department to measure company emissions for the 2021 base year.
I’m responsible for estimating Range 3 emissions Distribution chain.
That’s why I implemented the approach described in the article linked above.
After establishing the baseline, Restore the target Defined as a clear deadline.
For example, by 2030, your management can be reduced by 30%.
The role of the supply chain sector is then to design and implement plans to reduce CO2 emissions.
In the example above, the company reduces the reduction by 30% by using initiatives within the scope of manufacturing, logistics, retail and carbon offsets.
To support this journey, we developed analytical products to simulate the impact of different programs and help teams design the best sustainability strategies.
So far, these products have been in the form of web applications and have a user interface and are connected to the backend of their data sources.
Each module provides key insights to support operational decisions.
“Based on output, we can achieve 32% bulk emissions by relocating the factory from Brazil to the United States.”
However, for an audience not familiar with data analytics, interacting with these applications can still be overwhelmed.
How do we better support these users using AI proxy?
Agent AI for analyzing products
Now, we are developing these solutions by embedding autonomous AI agents that directly interact with analytical models and tools through API endpoints.
These agents are designed as Guiding non-technical users Throughout the journey, start with a simple question:
“How do I reduce emissions from the transportation network?”
Then, the AI agent is responsible for:
- Make the correct query,
- Connect to the optimization model,
- Explain the results,
- and provide actionable advice.
Users do not need to understand how the backend works.
They receive direct business-oriented output, such as:
“Implementing Solution XXX has an investment budget of YYY EUROS to reduce ZZZ TONCO₂Q emissions.”
By combining optimization models, APIs and AI-driven guidance, we provide an analysis-service experience.
We want to make sustainability analytics accessible to all teams, not just technical experts.
in conclusion
Use AI responsibly
Before closing, a sentence about minimizing the environmental footprint of the solutions we developed.
We are fully aware of the environmental impact of using LLM.
Therefore, the core of our products is still built on Deterministic optimization model,,,,, Our carefully designed.
Used only if the Large Language Model (LLMS) provides actual added values, mainly to simplify user interaction or automate non-critical tasks.
This allows us to:
- Guaranteed robustness and reliability: For the same input, the user always receives the same output, avoiding the random behavior of pure AI models
- Minimize energy consumption: Improve efficiency by reducing the number of tokens used in API calls and optimizing each prompt.
In short, we are committed to building solutions that are sustainable through their designs.
AI Agents Are Game Changeers in Supply Chain Analysis
For me, AI agents are becoming powerful ally to help our customers accelerate their sustainability roadmap.
This is a competitive advantage when I interact with a non-technical target audience as it enables me to provide analytical solutions that empower operations teams.
This simplifies one of the biggest obstacles a company faces when it begins a green transformation.
go through Communicate insights in simple language and Guide users to complete the journey, AI agents help Blink the gap between data-driven solutions and operational execution.
Let’s connect on LinkedIn and then twitter;I am a supply chain engineer who uses data analytics to improve logistics operations and reduce costs.
For advice or advice on analytics and sustainable supply chain transformation, feel free to contact me via Logigreen Consulting.
Samir Saci | Data Science and Productivity
A technology blog focused on data science, personal productivity, automation, operational research and sustainability…samirsaci.com
