What is EDIFACT?

The term UN/EDIFACT often also only "EDIFACT" is an abbreviation for the term: "United Nations Electronic Data Interchange for Administration, Commerce and Transport".

The term refers to an international standard format for electronic data interchange (EDI). The UN/EDFICAT standard has been published since 1988 and has since been regularly updated by CEFACT (a grouping within UNECE).

The EDIFACT standard is predominantly to be found in the European economic area. The corresponding format in the USA is ANSI X12.

New versions of the standard - also called Directories - are published every six months. The different versions of a year are then called e.g. 18A and 18B (18A: first version 2018, 18B second version 2018). An overview of all Directories can be found on the official website of the UNECE. The Directories contain the different message types that correspond to specific business transactions or documents. The most common message types are: ORDERS, ORDRSP, INVOIC, DESADV, SLSRPT and IFTMIN.

What is an EDIFACT file?

An EDIFACT file is basically like a simple text file that can be opened using a text editor. EDIFACT files often lack line breaks (CRLF) to minimize the file size. The EDI converter does not need you for reading in anyway, since each segment is terminated with a segment end character ('). The special thing about EDIFACT files is the biliterale or generally valid vote about the structure (syntax) of the file.

Simplified: With EDIFACT files, everyone knows in advance what must or could be in which position. The structure of the file is defined.

The following list gives an overview of all EDIFACT message types (UNSM) published by UNECE:

EDIFACT subsets

In addition to the Directories published by UNECE, there are also numerous subsets. Subsets are shortened directories adapted to the message types standardized by UNECE.

The message types for the various subsets are adapted to the requirements of the respective industry (subsets may not contain any new segments or changed segment sequences). A subset should always be readable with the higher-level UNECE standard format). The best known subset is the EANCOM subset of GS1. The EANCOM subset contains the post-item types and adaptations necessary for the consumer goods industry/trade. There are also subsets in other industries, which are often maintained by industry associations or interest groups.

The following EDIFACT subsets exist (DACH):

• CEFIC - Chemical industry
• EANCOM - Consumer goods industry
• Edi@Energy - Electricity and Gas (only valid for Germany)
• EDIBDB - Building materials industry
• EDIFICE - Electronics, software and telecommunications industry
• EDIFOR - freight forwarding industry
• EDIFURN - Furniture industry
• EDIGAS - Gas Transmission and Trade
• EDILEKTRO - Electrical Industry / Electrical Wholesale
• EDILIBE - Booksellers
• EDIPAP - paper manufacturer / paper wholesaler / paper converting industry
• EDITEC - Sanitary industry
• EDITEX - Textile industry
• EDITRANS - Transport economy
• EDIWHEEL - Tyre and wheel manufacturer (incl. AdHoc EDI)
• ETIS - Telecommunication (only for invoice)
• ODA/ODIF - General document formats
• ODETTE - Automotive industry
• RINET - Insurance industry

Structure of an EDIFACT message (explanation)

An EDIFACT message consists of the following syntax elements (according to ISO 9735).

Segments

There are two segment types:

• Service segments (starting with UN) UNH+1+ORDERS:D:18A:UN'
• User data segments (All other) e.g. DTM+4:20181224:102'

Service segments are easy to identify because they always begin with the two letters "UN". Service segments include for example: UNA (character definitions), UNB (header information), UNH (message header and identification of the message itself), UNT and UNZ (message close).

However, there are also the following less known service segments:

The user data segments contain the actual user data of the messages. i.e. values such as amounts, addresses, quantities, etc. Examples of known user data segments are: DTM (dates), RFF (references), QTY (quantities), NAD (addresses), PRI (prices), and many more.

Segment groups (SGX)

In EDIFACT there are segment groups in addition to the segments themselves. A segment group consists of at least two different segments. As an example: The segment group with the name SG29 in an ORDERS consists of the segments LIN & QTY. The segments, as well as the group, can be optional or necessary. The segment group can often repeat itself or the segments contained in it.

These two lines with the segments LIN and QTY are child elements of segment group 29. The segment group itself may repeat 200,000 times. The purchase order could therefore contain 200,000 different items with different quantities.

data elements

Data elements represent the smallest unit of an EDIFACT message. They contain "simply" data and have the following properties: length (e.g. 0-35) and occurrence (e.g. optional or mandatory). The following segment contains the data element C507 → 2380 from the segment DTM.

The date is displayed in 102 format. The value '102' (data element with code list - see next section) is in a code list and means that the date in this segment: '20181224' is formatted according to the format YYYYMMDD. However, the data element C507 → 2380 itself contains only the date.

Data element with associated code lists

If possible, data elements to which code lists are assigned should only contain values from the code list. A code list is a list with key-value pairs. For example, certain codes/abbreviations are assigned meanings. An example is the Incoterms in logistics: The code "EXW" means EX Works.

In practice, however, it often happens that free texts are also filled instead of the permitted values. Modern EDI converters can supplement code lists or accept any values and warn if the transmitted codes are not contained in the code list. This is not really recommended or intended.

In this case the code means "AIP" from the code list: A free text question. This means that the segment that follows contains a question: Delivery until Christmas possible?

In this case the code AIP from the code list means: A free text question. This means that the segment that follows contains a question. Namely: Delivery until Christmas possible?

Composite & Composite Data Elements

A "composite" contains several composite data elements. Thus a composite can also be referred to as a group of data elements within a segment.

An example of a frequently used composite is the composite C507 in the DTM segment. Here, three composite data elements (2005, 2380 & 2379) are combined in one composite.

Separating and control characters

In a message, the various elements are separated by separators. Essentially, these are the data elements and the composites. Which characters are used to separate the message is defined in the UNA service segment. In the standard definition, the plus sign (+) is used to separate data elements & segments and the colon (:) is used for composite data elements.

Another important control character is the segment end character, which is a single apostrophe in the standard system. This character is used to close the segment or initiate a new one. In the productive operation of an EDI system, the line breaks (CRLF) are omitted in favour of the file size in the EDIFACT message.

Example of an order in EDIFACT (EDIFACT ORDERS)

The following EDIFACT file shows an order sent via Edifact from 01.12.2018 to Amazon (GLN: 4062268000006). The sender of the message is the company menten (GLN 4399901544994). A Christmas tree (1st piece) is ordered with the order number ORD4711.

The definition/directory or the EDIFACT message itself can also be easily displayed using a EDI-Viewer.