Skip to main content

Chemical Products : Processing guide

S
Written by Support

1. Align with the client’s specific requests

Clients might have specific requests for data processing → they will have an impact on which module format you should opt for.

Suppliers or other custom EFs (outside of Greenly’s database)

  • The client must have paid for their integration, and the EFs should have been validated by the R& M team.

  • These EFs must be added to the EF Manager as company-specific EFs before you can retrieve their IDs.

For large accounts

  • The best is to validate the EF association to the client’s data with the client → the advanced analysis need to be used for that.

2. Data Processing

The goal of an activity-based study is to categorize a client’s data, meaning associating each activity of a specific category to the most suited emission factor.
If you are working on a multi-year account, don’t forget to check the Rebaselining section.

Manual processing (Advanced module)

How to extrapolate missing data

  • Missing Name:

  • Fill with unknown and use the category (if provided) to match the EF.

  • Missing / wrong Unit:

  • Contact PM.

  • Missing Category / Section:

  • Try to replicate the proportion of section in the existing data.
    For example, if the existing data has (as % of total weight):
    30% “Wood Products”
    30% “Plastic Products”
    40% “Electronics”,
    then your extrapolation on the missing sections would have the same proportion.
    So if 10 lines have missing section, 4 of them would then get “Electronics”, 3 of them “Wood Products”, 3 of them “Plastic Products”. For small dataset, doing it manually is fine. For big ones, make a script.

  • Missing Quantity or Unitary Weight:

  • Most of the time, when quantity is left blank but UNITARY WEIGHT is huge (or the other way around), then it’s probably because quantity is 1.

  • If some quantities / weight have values, but a few other don’t: that’s missing data. Then you have two choice:

  • Long deadline:

  • Ask PM for more info & contact client.

  • Short deadline:

  • Average the quantity of the whole dataset (imprecise but methodologically ok) to extrapolate the missing quantities.

  • Average the quantities of the category / section and apply it to all row that have this section but missing quantities.

  • If data has multiple missing field:

  • Order to follow will be:

  • Extrapolate Name (Unknown)

  • Extrapolate Section

  • Extrapolate Weight

  • Extrapolate Quantities

3. Rebaselining

The goal of rebaselining is to apply the latest version of EFs to the base year and to ensure that categorization is harmonized with the current year → This way you will be able to study the annual evolution of emissions.
💡 When you do an analysis for a company, you should first check if a similar analysis has been done for the previous years, and check with the deal owner/client if a rebaselining of the data has been planned. It’s very important to make sure that, across years, similar modules were used. If, for example, your base year contains an Advanced module (e.g., Advanced version of the Freight module) and the current year has the end-to-end Freight module, then it’s best to align in the base year and use the end-to-end Freight module instead of the Advanced one. Having consistency in the modules used across years ensures that the EFs and methodology used are consistent over years, which means that you can compare emissions.
⚠️ Do not proceed to any rebaselining without aligning with the deal owner/your client on the specifics you must take into account (custom emission factors, EF exclusions that differ from the current year, PC override, etc.)
For more information about rebaselining in general, you can check:

🅰️ Automatic rebaselining

The algorithm of the E2E module takes into account previously done studies to adapt its categorization suggestions → You can upload the data and follow the module’s process for an efficient rebaselining.

Ⓜ️ Manual rebaselining

👉 Avoid using this section unless the data you have to rebaseline is very little.
🎯 Follow the below steps to process the data:

  • [TO BE POPULATED]
    💡 Try to harmonize categorization throughout the years as much as you can.

Related Articles
Did this answer your question?