Sequence recovery is the structured process by which a courier hub’s control systems detect, re-align, or abandon the planned order of parcels between two identification points. A route describes where a parcel should go; a sequence describes when a parcel should arrive relative to its neighbours. In high-rate sortation systems, the WCS maintains a logical mission list, the PLC maintains a physical slot map, and the gap between the two is where recovery logic operates. This article explains how that relationship works, what happens when it fails, and where the boundary lies between automatic recovery and human decision-making. It is written for maintenance engineers, controls teams, and shift operations staff who need a common language for diagnosing sequence-related sortation faults.
The Role of Sequence in Hub Sortation #
Most courier hubs do not sort parcels by reading their barcode at the destination point and then diverting them. Instead, the system reads the parcel early, records its identity and destination, and then issues a divert command later, timed against the parcel’s expected arrival at a chute, lane, or takeaway conveyor. This timing is only trustworthy if the physical order of parcels matches the logical order stored in the WCS.
Sequences exist at several levels: between the induction scanner and the first merge, between a merge point and a downstream scanner, and between a secondary scanner and the destination chute. Where no scanner exists downstream, the PLC relies entirely on slot tracking – a fixed mapping between conveyor position and the parcel’s logical identity. When that mapping breaks, the system cannot reliably decide where to send a parcel.
Recovery is therefore not a single feature. It is a set of rules that decide what to do when an expected parcel does not appear, an unexpected parcel appears, or two parcels arrive in the wrong order.
What Breaks a Sequence #
Sequence breaks are caused by physical behaviour that contradicts the planned order. The most common causes in a courier depot environment include:
- Double induction, where two parcels enter the same carrier slot or pass the induction scanner inside one minimum gap.
- Short-spaced parcels that violate the minimum pitch, causing the PLC to count one slot where two parcels physically exist.
- Undetected jams or momentary stalls that delay a parcel without tripping a jam alarm immediately.
- No-reads, where the scanner fails to decode the label and the parcel continues without updating its logical identity.
- Misreads, where a damaged or partially covered label produces a wrong but valid-looking code.
- Recovery actions already consumed upstream, such as a merge gate that released a gap and thereby shifted slot timing.
- System restarts, which can reset encoder counts or slot pointers while parcels remain on the conveyor.
A sequence break does not mean a parcel is lost. It means the system can no longer prove which physical parcel corresponds to which logical mission at a given decision point.
Component Interactions and Recovery Windows #
Three control layers interact during normal sortation and during recovery. The WCS holds the mission list: every parcel’s identifier, destination, and planned path. The PLC holds the slot map: a time-synchronised representation of what should be at each encoder position on each conveyor segment. The field devices – scanners, photoeyes, encoders, and diverters – provide physical confirmation of what actually moved.
The critical concept is the recovery window: the conveyor distance, expressed in time or encoder counts, between a confirmation point and a decision point. If a confirmation scanner reads a parcel and the next decision point is far enough away, the WCS has time to reorder its mission list and issue a corrected divert command. If the window is too short, the parcel must be allowed to pass the decision point and be routed to a recirculation loop or an exception lane.
Recovery Algorithm Types #
Three general approaches are used, often in combination. Compare-and-correct logic records the expected order of labels and compares it to the actual order reported by a downstream scanner. When a mismatch is found, the WCS updates the mission list for all parcels between the mismatch point and the scanner. Slot-shift logic operates in the PLC: when a gap sensor detects missing or doubled occupancy, the PLC shifts all downstream slot numbers by one or more increments. Logical reordering does not move any physical conveyor; it simply changes which destination is assigned to each upcoming divert command.
Each approach has limits. Compare-and-correct requires a downstream identity confirmation point. Slot-shift requires accurate gap sensing. Logical reordering requires that the physical conveyor has not already passed the decision point.
Observable Symptoms #
Operators usually notice sequence problems through indirect symptoms rather than a direct alarm. Common observations in a courier hub include:
- Recurring “unexpected scan” alarms, where a scanner reads a parcel label that the WCS did not expect at that position and time.
- Destination lanes that consistently short-count, even though no parcel was seen going to the wrong chute.
- Merge rate collapse, where the system throttles induction because downstream recovery logic is waiting to resolve a mismatch.
- “Phantom parcel” alarms at the end of a shift, where the WCS believes a parcel is still on a conveyor or in a lane but no physical parcel can be found.
- Divertors firing on empty cells, which indicates that the PLC’s slot map believes a parcel is present when the physical conveyor is empty.
- Repeated no-read or re-scan loops on the same parcel, which suggests the parcel was diverted into a recirculation path because the system could not trust its sequence.
Evidence Collection for a Sequence Event #
Diagnosing a sequence break requires correlated evidence, not just one screen or alarm log. Collection should begin immediately after an event, before buffers are flushed or shift counts are reset. The following evidence is typically needed:
- Scanner read logs with timestamps and confidence scores for all read points between the induction source and the affected destination.
- Photoeye transition histories with millisecond timestamps, particularly for gap sensors and merge in-feed sensors.
- Encoder counts for the affected conveyor segments, captured before and during the event.
- WCS decision logs showing which divert command was issued for each logical mission.
- PLC slot table snapshots, which show the expected occupancy at the moment of failure.
- Video footage of the induction area, the merge point, and the affected destination approach.
Time alignment is critical. WCS logs, PLC logs, and video time stamps may be sourced from different clocks. A diagnostic workflow should include a time-synchronisation check before any comparison is made.
| Symptom | Likely Area | Evidence to Collect |
|---|---|---|
| Consecutive misdiverts to the same lane | Merge timing or WCS mission order | WCS decision log, merge photoeye timestamps, destination lane counts |
| Occasional “unexpected barcode” alarm | Induction double-feed or upstream no-read | Induction scanner logs, gap photoeye data, video of induction in-feed |
| Divertor fires on an empty cell | PLC slot shift after gap detection | PLC slot table snapshot, encoder counts, photoeye state history |
| Gradual rate drop after restart | Recovery window insufficient or slot pointers reset | Merge gate open times, WCS throttle commands, restart records |
| Phantom parcel at end of shift | Logical mission never closed | WCS mission status, destination scan confirmations, lane flush records |
Common Interpretation Errors #
A sequence fault is frequently misdiagnosed because its symptoms resemble a scanner, mechanical, or communication problem. One of the most common errors is treating a logical reorder as if it were a physical re-sequencing activity. The WCS can change the destination list in its memory, but it cannot move a parcel that is already past its divert point. Assuming otherwise leads engineers to search for a logic fault that does not exist.
A second error is assuming a “phantom parcel” means a parcel physically exists somewhere on the conveyor. In most cases, the phantom is a logical mission that was never confirmed as delivered, cancelled, or recirculated. The parcel may have already been dispatched or may never have existed in the first place.
A third error is blaming the scanner when the root cause is slot drift. A scanner reads exactly what is in front of it; if the PLC slot map has shifted, the scanner output is correct but the expected order is wrong. Replacing or cleaning the scanner will not resolve the fault.
Finally, teams often overlook that recovery has a finite budget. If the upstream logic has already performed several reordering operations, the system may conservatively route parcels to exception lanes even though the physical conveyor appears normal. This is not a fault; it is a protective decision.
Maintenance Implications #
Recurring sequence breaks are often a maintenance signal rather than a controls fault. Physical conditions that erode recovery reliability include contaminated scanner windows, stretched or slipping belts that skew encoder-to-distance relationships, worn photoeye brackets that change sensing position, and divertor timing drift caused by air-pressure fluctuation or mechanical wear.
Preventive maintenance should include regular verification of encoder calibration against a known distance, cleaning of scanner windows and photoeye faces, and checking that gap sensors trigger at the correct moment relative to parcel edges. These checks are especially important after conveyor modifications, belt replacements, or controller swaps, since a small change in physical timing can consume most of the recovery window.
Trend analysis of “unexpected scan” and “gap violation” alarms can identify degrading conditions before they cause visible missorts. A gradual rise in these events over several weeks often points to mechanical drift, while a sudden spike is more likely to be a configuration change or a specific physical obstruction.
Decision Boundaries #
Automatic recovery is designed to handle bounded mismatches. It is not designed to make the final decision when a parcel is unaccounted for at a safety-critical or dispatch-critical boundary. The boundary between automatic recovery and human intervention should be explicit at each site. It depends on the recovery window length, the presence of a downstream confirmation scanner, and the consequences of a misdivert into a dispatch trailer.
Operations teams should know which actions require a competent engineer or controls specialist. Examples include resetting slot table pointers, manually flushing a lane, re-baselining the WCS mission list, or declaring a parcel lost for audit purposes. These actions should only be carried out in accordance with site procedures, OEM documentation, and local lockout and safety requirements. Nothing in this article authorises bypassing a safety device or overriding a safety-related interlock. Site procedures and competent engineering judgment always take priority.
Decision boundaries are not fixed. A recovery rule that works at low rate may fail at peak rate, because the time available to reorder missions shrinks as the physical gap between parcels decreases. Shift teams should review sequence alarm trends after rate changes, not just after mechanical failures.
Key Takeaways #
- A sequence is a time-based order, not a destination route. Both must be correct for a clean sort.
- Recovery is bounded by physical windows; the WCS can only reorder what it can confidently re-identify before a decision point.
- Evidence must be time-aligned across WCS, PLC, scanner logs, and video before drawing conclusions.
- Phantom parcels are usually logical missions that were never closed, not parcels lost inside the conveyor.
- Recurring sequence breaks are often caused by mechanical drift or encoder inaccuracy, not by controller logic.
- Scanner replacement will not fix a slot-shift fault; confirm the slot map before changing field devices.
- Automatic recovery has a finite budget and may deliberately route parcels to exception lanes when mismatches accumulate.
- Always follow site procedures, lockout requirements, OEM documentation, and competent engineering judgment; never bypass safety devices.
