How do you build an FDES configurator? The project stages

When people picture an FDES configurator, they think of the screen first: some fields, a button, an FDES coming out. Yet it is the last thing to be built. Before the interface, you need to understand your products, know what makes them vary and translate that into an LCA model. You also need to understand your processes, to find the best way to fit the FDES configurator into your organisation (automatic creation when a quote is entered, access for field sales teams, shared access across your teams). This guide follows the project in the order it unfolds, and sets out at each stage what you will need to prepare.

What are the stages of an FDES configurator project?

A configurator project runs in five stages: scoping the project, collecting and structuring the data, building the LCA model, developing the application, then testing it and getting it verified. The first three happen upstream, before a single line of code is written.

The order is not negotiable. Every field shown on screen corresponds to a variable in the LCA model, with its allowed values: an interface developed before the data and the model are stable will have to be reworked.

What data needs to be prepared?

There are two kinds of data to prepare: the data that builds the LCA model, gathered once, and the data the user will enter for each FDES. Where you draw the line between the two decides what will be able to vary.

Missing or approximate data does not always block the project, but it must be identified at the scoping stage. A project that makes sense on paper may first require real work on the data: it is one of the five factors in the suitability guide.

How does the LCA model become an application?

By translating each element of the model into an element of the application: variables become fields, ranges become checks, calculation rules become the engine, and deliverables become exports.

You then have to prove that the application calculates correctly. This is done first with test configurations whose results are known, then through verification: an independent verifier, approved by the INIES programme (INIES is the French national EPD database), examines the model and its translation into the application. The verifier is independent of every party to the project: neither the manufacturer, nor the LCA practitioner, nor the company that developed the configurator. INIES publishes the list of approved verifiers, and the verification guide details what they examine.

How do you make the tool simple for non-experts?

By starting from what users know, not from what the model contains. A sales representative knows the product reference ordered and the construction site; a product engineer knows the formulation. Neither of them should have to handle an emission factor.

In practice, this means forms written in the language of the trade, step-by-step input in the order users think about their product, drop-down lists rather than free text, and error messages that say what to correct.

This is the approach taken by the DimACV configurator, designed with the CTMNC for the natural stone sector: manufacturers, often not LCA experts, describe their quarry, their processes and their formats, and the project’s methodological rules apply automatically. The DimACV project shows this user journey.

What features should you plan for?

It depends on who will use the tool, and for what. Six features come up often in projects.

What can complicate an FDES configurator project?

The main difficulty in an FDES configurator project is usually collecting the data needed for the environmental modelling, and its quality. Other factors can also affect its complexity.

• Data that is hard to collect: information on raw materials, energy consumption or manufacturing processes may be spread across several departments, production sites or suppliers.
• Very different products: some product families have compositions or manufacturing processes that require several LCA models.
• Parameters that are hard to define: you need to identify the characteristics that can vary and their limits of variation, in line with the INIES programme rules specific to the type of configurator chosen (generator or distributor).
• Data of variable quality: some information may be incomplete, not very representative or insufficiently documented.

The scoping stage identifies these issues, assesses the data available and defines a scope suited to the project.

How do you prepare an FDES configurator project?

Before starting to develop an FDES configurator, it is worth taking stock of the products concerned, the data available and the project’s objectives. Here are five points to review to make scoping easier.

1. 01Identify the products concernedWhich product families do you want your FDES configurator to cover?
2. 02Estimate the number of FDES to produceHow many product references or variants do you want to be able to declare?
3. 03Check whether you already have LCAs or FDESThey can serve as a starting point for your project.
4. 04Identify the data availableDo you know where your manufacturing, composition and transport data is?
5. 05Clarify your objectivesDo you want to produce FDES in-house, integrate them into your tools or let your customers customise certain declarations?

PDF · 1 page

Get this checklist by email

Key takeaways

An FDES configurator is both an LCA project and a software project. It has three prerequisites: products you know how to describe, data you know where to find, and a use for the FDES you can put into words. The rest is built with a team that combines both skill sets, preparing for verification from day one.