Central-fill pharmacy operations are replacing or augmenting traditional sorting systems with Automated Guided Vehicle (AGV) fleets, with a 2024 two-stage assignment study quantifying the trade-off across throughput, utilization, and cycle time KPIs [S1].
The decision sits between fixed conveyor sorting lines (high throughput, lower flexibility) and AGV-based cells (higher flexibility and scalability, higher orchestration complexity) for distribution centers handling several hundred to several thousand SKUs under expiry-date controls [S1][S2].
What the 2024 AGV Optimization Study Established
A two-stage package assignment model applied to central-fill pharmacies, implemented via discrete-event simulation and a simulation-based heuristic, was published 2024-08-19 in The International Journal of Advanced Manufacturing Technology, volume 134, pages 2439–2457 [S1]. The model jointly optimizes the sorting-area assignment and the upstream traffic flow, an interaction prior work had treated separately [S1].
Performance was measured against three KPIs: throughput, AGV utilization, and cycle time, with a sensitivity sweep on the number of AGVs, demonstrating that assignment strategy can move the system efficiency as much as adding or removing vehicles [S1]. This is the core engineering reason AGV projects fail when teams only adjust fleet size without re-engineering the assignment logic.
Conveyor Sorting Lines vs AGV Sorting Cells: A Criteria Comparison
Selection reduces to four engineering criteria: throughput ceiling, layout flexibility, peak-load handling, and integration cost with a power distribution and WMS/MES layer. The table below summarizes the documented behavior. [S1]
Conveyor sorting lines: high sustained throughput, fixed mechanical path, low per-package variable cost at steady state, harder to scale in steps. AGV sorting systems: throughput is a function of fleet count and assignment logic rather than belt length, with the 2024 study showing meaningful cycle-time reduction under the two-stage heuristic versus naive routing [S1]. For a related e-commerce throughput-tiers and sorter type match-up, see the E-commerce Sorting System Selection guide.
For a distribution cabinet and power distribution box retrofit, AGV charging bays add three-phase loads and harmonics filtering that fixed conveyors typically do not, which is often the most under-budgeted line item.
Who This Selection Is For, and Who It Is Not For
AGV sorting systems fit central-fill pharmacies and large hospital distribution hubs running thousands of unit-dose or multi-dose prescriptions per shift, where expiry-date, lot, and storage-condition rules (cold chain, controlled substances) drive traceability demands similar to those applied during pharmaceutical supplier evaluation [S2]. They are a poor fit for small retail pharmacies below ~500 scripts per day, where the orchestration overhead of an AGV fleet outweighs the throughput benefit.
Conveyor lines remain the correct answer where the SKU count is moderate, the product mix is stable, and the facility has long straight runs available; they are also the safer answer in explosion-risk environments where the AGV battery/charger zones would force additional explosion-proof distribution hardware, raising installed cost significantly.
Selection Criteria a Process Engineer Should Score
Engineers should score at least five quantitative criteria before committing to a topology: peak daily package volume, mean and 95th-percentile cycle time targets, AGV utilization band (the 2024 study reports the heuristic improves this KPI directly) [S1], upstream station dwell time, and the WMS/MES integration surface area. Each criterion maps to a measurable KPI from the 2024 study, which is what makes the paper a usable reference rather than a marketing piece [S1].
Lot-level and expiry-level traceability should be scored separately, since both conveyor and AGV systems must interface with barcode/RFID capture, but the latency budget differs. On a high-speed conveyor, vision-and-scan dwell is on the order of milliseconds; on an AGV cell, the package can pause at an induction station, which is a benefit for multi-step verification but a penalty for pure throughput.
Failure Modes and Constraints in Pharmaceutical Sorting
The dominant failure modes in pharmaceutical sorting are mis-sort under high SKU churn, AGV deadlock at intersections under unbalanced flow, and expiry-date mis-prioritization during peak load. The 2024 study explicitly identifies the upstream traffic flow as the lever that, if left unmodeled, makes the sorting-area assignment strategy underperform in simulation [S1].
A second constraint is the regulatory and waste-handling layer: discarded and spent pharmaceutical items must be sorted into segregated containers per applicable rules, a system architecture that pairs input devices (barcode, RFID, handheld) with a control system that opens only the assigned container lid, documented in US patent US7311207B2 granted 2007-12-25 [S3]. Any new sorting system should be designed so that rejected, recalled, or expired units can be diverted into a parallel segregated stream without re-engineering the main sort logic.
Standards, Integration, and Sourcing Notes
No single ISO or IEC standard governs "pharmaceutical sorting system" as a complete assembly; instead, the build references GMP-aligned traceability, electrical codes governing the power distribution box and any explosion-proof distribution zones, and supplier-qualification frameworks of the type analyzed in multi-criteria pharmaceutical supplier sorting models under uncertainty [S2]. Teams should request the supplier's KPI evidence in the same three-axis format (throughput, utilization, cycle time) used in the 2024 study, which keeps vendor claims comparable [S1].
Two trackable signals to watch through the rest of 2026: vendor publication of two-stage heuristic parameters in commercial RFQs, and the appearance of central-fill pharmacy RFPs that score AGV assignment strategy as a separate line item, which would indicate the 2024 methodology is moving from research into procurement practice [S1].