The FMS predicts from the data it has

Why does a valid-looking route still need checking?

A computer can execute precisely the wrong data.

Different databases support different functions

A navigation database contains coded airports, fixes, facilities, routes and procedures appropriate to the installed system. It is associated with an effective cycle and must be checked against the planned operation and applicable procedure. AIRAC commonly uses a 28-day cycle, but an active and a standby database are not simply two interchangeable route memories. Activating a different database can affect stored or entered data according to the installation.

The performance database represents a defined aircraft/engine configuration and supports predictions and optimisation. It is different from the navigation database. A cost index influences the time-versus-fuel trade-off within the system’s logic; a performance factor can adjust predictions for aircraft performance condition. Neither input authorises ignoring a structural, speed, fuel or operational limitation.

Verify identity, position and route together

A typical setup checks aircraft and engine identity, database applicability, position initialisation, route, runway/procedures, mass and performance inputs. Mandatory boxes and optional fields have different meanings; an empty optional field is not evidence that a required value is correct. The scratchpad is an input or message area and does not by itself prove a value has been inserted into the intended field.

Review the route geometrically and against the clearance, including discontinuities, constraints and similarly named fixes. A dual FMS may share data through a cross-talk link; two identical displays may therefore contain the same entered error. A failed CDU can remove one interface while its computer remains available, whereas a computer or cross-talk failure has different consequences. The laboratory preserves the route but changes forecast wind, showing that a coherent displayed plan can still have a materially different prediction.

Why can an FMS fuel prediction be internally correct and operationally wrong?

A prediction is conditional on its model and assumptions.

Combine planned and measured data

The FMS can combine entered mass, winds, route and constraints with measured position, speed, altitude and fuel-related information. Its present position may be selected or blended from inertial, GNSS and radio-navigation sources. A reduction in source availability can change accuracy or integrity and affect the capability to conduct a particular navigation operation. A displayed position is an estimate with a source architecture, not a physical landmark.

Time and fuel predictions depend on the route remaining, the expected groundspeed, the planned vertical profile and the performance model. A new wind, vector, level restriction or actual fuel-flow difference can make the prior estimate obsolete. The computer may continue producing smooth numerical predictions even when a damaged or abnormal configuration is not represented in its model.

Validate the prediction against the flight

Compare expected and actual fuel consumption at meaningful points and look for a trend rather than waiting for a destination number to cross a limit. Check whether the model recognises engine-out, gear-down, flap, speedbrake or other relevant conditions. A single correction to the destination estimate is not a substitute for understanding why actual use differs.

Raw navigation measurements provide a useful cross-check when the displayed FMS path and radio guidance disagree, subject to procedure and equipment limitations. In the laboratory TAS, wind and remaining distance generate time, then fuel flow generates trip fuel. Increase headwind and watch both time and fuel rise together; increase flow separately and observe fuel change without route-time change. This preserves the causal relationship and avoids presenting the prediction as a guaranteed outcome.

Can you explain and apply the mechanism?

Four original questions. Work from the stated facts, then compare your reasoning with the explanation.