Shared reference: define the operational meaning
Start by describing shared reference through an example from your operations.
Rail integration needs to connect a service with its units, locations and events. Participants may identify the same operation differently. Before transforming messages, agree which combination recognises the movement in each system and what references the partner should return to preserve that relationship across the exchange.
Service units: establish the actual scope
Reviewing service units separates available functions from remaining work.
Identify the source of every update: operator, terminal or another participant. A planned departure does not mean a confirmed event. The receiving system needs the information's operational meaning and should not infer execution from a value that the source uses only for planning or coordination.
Event authority: identify evidence of completion
For event authority, distinguish sending information from actually using it.
Units may be reassigned or appear in different services. Design should use the context needed to recognise the intended operation rather than rely only on a unit identifier. Ask how reassignment is reported and what prevents an update attaching to an earlier service involving the same unit.
Service information should stay connected across operators and terminals
Teams need to match services, units, locations and events even when participants use different references. Composition changes need an explicit relationship to the earlier service. Check whether the reference remains or a new one is created and reported. Integration should follow that operational decision rather than infer continuity solely because some units happen to be shared between the previous and revised service information.
What a proposal should resolve for rail operations
Describe the operators, terminals and customers involved and what your system must do with their information. Review scope: EDI integration for rail transport.
An example to frame the decision
Illustrative scenario, not attributed to an actual customer. A rail operator exchanges information with terminals and customers using different references. Staff match services and units to update planning and communicate milestones.
Map references, associate units with services and adapt messages to participant systems. Separate forecasts from executed events and prepare agreed responses. Reassigned units, unknown references and service changes need agreed authoritative sources and update confirmation ownership.The example helps review the complete journey and integration decisions; it does not imply an already available connector or measured savings. Use it to define a comparable need in your own business.
Dispatch needs usable orders and statuses returned to the customer
TMS integration follows the service: requester, collection and delivery points, customer reference and possible changes. Test a planned service, changed assignment and executed event. Check references at both ends after each step. If an event arrives before the information needed to relate it, define its pending treatment and review owner so the connection does not create an incorrect operation simply to complete processing.
Create a service with the information it needs
Review the data the TMS requires to register an order: references, parties, collection and delivery points and other applicable information. Missing data needs an agreed enrichment or review process.
Compare proposals using the same workflow
Give each provider the same incoming document, intended result and required response. Ask how it covers shared reference, service units and event authority. Compare partner adaptation, existing-software tasks and acceptance criteria. Identify the effect of adding documents or changing specifications. A lower price may cover fewer stages; a broader project may include functions you do not yet need. Compare useful scope before comparing price, so the decision reflects the workflow your team actually needs.
Prepare the first discussion
Identify software name and version, partner and available documentation. Collect a representative service, associated units and expected event sequence for assessment. Identify who confirms each milestone and which references the destination requires. These inputs support an operational rail connection instead of a collection of individually valid messages whose relationships cannot be verified across the participating systems.
Where Ediprism fits
Ediprism is EDI integration software between your business systems and third-party applications. Ediprism can connect rail exchanges to existing system interfaces, adapting references and responses to project requirements. Access to every platform and equivalence between service models should not be assumed. Scope must confirm the operations that each participating application can actually receive or issue through an available interface.
Initial cost, continuity and later changes
Separate implementation, third-party services and maintenance. Separate partners and flow directions in the quote. Receiving one operator's plan and returning confirmations to another may require different rules. Identify each participant's data contribution and dependencies involving rail software, a TMS or the terminal, rather than pricing all messages as one undifferentiated exchange.
One service, several organisations
Operators, railway undertakings, terminals and customers may use different references and applications. The project must identify which part of the service each datum represents and who can authoritatively communicate an update.
Match units, services and locations
Translating a field is not enough. We need to check unit identifiers, service relationships and the meaning of locations and statuses in both systems. Rules are agreed with partners to avoid records that arrive but cannot be used.
Events that can actually be communicated
We review which milestones are recorded in the system and which require additional information. A SIMPLE connection is scoped separately when needed; it is not inferred merely because a business operates rail transport.
What we review before quoting
Identify existing software, the partner and its documentation, inbound and outbound flows and available interfaces. This enables agreement on test scenarios, exception ownership and readiness criteria. The proposal should reflect adaptations needed for the specific exchange.
Return statuses from confirmed events
The status sent to your customer must correspond to an agreed TMS milestone. Codes are mapped between systems and rules define updates, corrections and destination rejections.
One TMS, multiple customers
Distinguish the system data model from shipper-specific requirements. This makes it possible to quote what can be reused and which adaptation is needed for a new partner.
What we need to assess the connection
Software identity and version, available interface documentation and an anonymised example exchange. Permissions, test environments and responsibilities are reviewed with the vendor or developer. Connectivity is confirmed by scope; brand compatibility is not claimed without verification.
A first flow with standalone value
For a limited phase, define one document, one partner and one useful operational result. For services crossing several locations, specify the location behind each date and event. The same status label can mean different facts depending on the participant. Tests need that context so a terminal milestone is not interpreted as completion of the entire service by the receiving application.
References that help investigate exceptions
Users reporting differences need to locate the corresponding message and operation. Support should establish which participant reported each event and under which reference. This helps distinguish planning changes from execution discrepancies. Agree how corrections are communicated so a technical message history supports, rather than replaces, the operational team's interpretation of the service and its actual movements.
Distinguish objectives from verified capabilities
A business need describes the result sought but does not prove the system has every required function. At transition, identify ongoing services and already reported events. Decide how each continues through the new route. Enabling exchange without this review could resend historical information or leave an open operation without the reference its recipient needs to recognise subsequent updates from the connected systems.
Decide the next step
Identify the unresolved question and information needed to answer it: perhaps an ERP interface, amendment meaning or customer response. Handover should show event authority and keys connecting each event with the service. This supports participant disputes without automatically treating the last received message as correct. Rail coordination needs provenance and meaning as well as field transport so teams can understand the operational history behind an update.
1. Identify the exchange
Capture the business need, partner specification and existing system. Confirm the result each document should produce and the data it requires.
2. Agree the solution and ownership
Define interfaces, mappings, validation and responses. Responsibilities across Ediprism, your team, software vendor and partner must be documented alongside prerequisites.
3. Test realistic scenarios
Agree inbound and outbound examples, amendments, retries and exceptions. Testing verifies final-system data and partner responses. Timing also depends on access and third parties.
4. Prepare go-live and continuity
Determine monitoring ownership, incident escalation and maintenance before operation. Version and specification changes need a review, testing and acceptance process.
A module in your existing software
A module may fit if it covers required documents, partners, versions and business rules. Ask what is included and what needs configuration or development. Check the effect of adding customers or changing specifications, since fit depends on scope and maintenance.
Custom integration development
This can fit when your team intends to build and maintain the workflow and understands the systems and partner requirements. Include interpretation, mapping, responses, tests and subsequent changes. The first transmission is not the whole project: identify incident ownership and documentation responsibilities.
An integration layer such as Ediprism
This is an option when you want to connect partner formats to existing systems while separating exchange adaptation from business logic. Assess Ediprism by message coverage, available interfaces and intended operational results. Scope should identify adaptations and ownership shared with your software vendor.
An exercise to evaluate the proposal before buying
Assessment scenario; not results from an actual customer.
Select a rail service with several units and different operator and terminal references. Explain the keys linking records beyond the unit number. Reassign a unit and check preservation of service context. Use a recognisable operational sequence rather than independent messages validated only for structure.
Add a forecast and an executed event from different participants. Check source authority and destination interpretation. Where an event precedes its matching information, establish pending treatment and review ownership instead of creating another operation or treating the forecast as proof of execution.
Request evidence connecting service, unit, location, event source and known response. Name TMS or other software responsibilities where involved. Compare the same sequence across offers without inferring access to every platform or support for every service-model variation from a general claim of rail integration.
Reference documentation
Partner and system specifications define the project. These references describe the framework, not Ediprism compatibility or certification.
Before you choose a provider
How do I apply this guide to my case?
Identify your software, partner and intended result. Assessment should review shared reference, service units, event authority using examples and available interface conditions.
How do I compare providers?
Provide both with the same document, system and intended result. Compare workflow coverage, interfaces, responses, implementation, support, changes and cost.
How much does integration cost?
Cost depends on partners, documents, interfaces, rules and tests. Separate implementation, adaptations, third-party costs and maintenance.