Managers: Stop Blocking and Starving with Five Zone Picking Fixes

Manager-ready fixes to eliminate blocking and starving in zone picking. Five low-friction tactics you can pilot in two shifts, backed by recent research.

Zone picking optimization pays off fastest when order volume is high and SKU overlap across orders is heavy. The single priority is workload balance: fix routing, batching, and slotting so no zone sits idle while another backs up. That imbalance, not the zoning concept itself, causes most throughput loss. Scaling the fix past one pilot zone will eventually require software that can route and rebalance in real time.

TL;DR:

  • Zone picking excels with high order volume and SKU overlap, particularly when workload is balanced across zones to prevent bottlenecks.
  • The main performance drain is zone handoff delays, which can be mitigated through adaptive routing, time-slot batching, and smoothing merge points.
  • Tracking throughput, blocking, starving, and merge counts with simple timestamps reveals operational bottlenecks without new hardware investments.
  • Short, focused pilots on small product groups using baseline measurements provide clearer insights before expanding changes system-wide.
  • Orchestration software significantly improves throughput and workload balancing by reacting dynamically to real-time zone loads, especially when manual fixes reach their limit.
  • Table of Contents

  • What Is Zone Picking and How Does It Differ From Batch Picking?
  • Zone Picking vs. Batch and Wave: What Actually Drives the Trade-Off?
  • Which Metrics Reveal Zone Picking Bottlenecks?
  • What Tactical Changes Improve Zone Picking Right Now?
  • Why Does Orchestration Software Change the Math?
  • How Do You Pilot This Without Betting the Whole Operation?
  • When Should You Fix the Process and When Should You Adopt a Platform?
  • Ready to Move Past Manual Zone Rebalancing?
  • Sources
  • What Is Zone Picking and How Does It Differ From Batch Picking?

    Zone picking assigns pickers to fixed sections of the warehouse. Each picker only touches SKUs in their zone, and orders move from zone to zone until every line is filled. That is fundamentally different from discrete picking, where one picker walks an entire order start to finish, or batch picking, where one picker collects multiple orders in a single pass through the whole facility. Zone picking trades picker travel for coordination overhead. You cut walking distance dramatically, but you introduce handoffs, conveyors, and timing dependencies that discrete picking never has to worry about.

    The main variants show up in almost every distribution center sooner or later:

  • Sequential (pick-and-pass): totes move zone to zone in order, one picker at a time, simple to run but vulnerable to blocking if one zone lags.
  • Parallel (pick-and-merge): multiple zones pick simultaneously and orders consolidate at a merge point, faster but merge-point congestion becomes the new constraint.
  • Dual-lane: two parallel conveyor lanes let one zone work ahead while the other catches up, reducing makespan compared to single-lane sequential designs in modeled scenarios.
  • Hybrid zone/batch: zones handle high-density SKU areas while batch picking covers slow-moving overflow.
  • High order volume with heavy SKU overlap favors sequential or dual-lane. Scattered SKU profiles or temperature-controlled sections often justify a hybrid approach instead.

    Zone Picking vs. Batch and Wave: What Actually Drives the Trade-Off?

    Choosing a method comes down to three questions: how big are your orders, how much do their SKUs overlap, and how tight is your shipping window. Get those answers first, because the wrong method choice compounds every other inefficiency you try to fix later.

  • Order size and SKU overlap. Zone picking wins when orders share SKUs across a limited number of high-traffic areas. Meal-kit and grocery-style operations, where the same handful of SKUs appears in nearly every order, are the textbook case, and routing research on sequential zone systems confirms that repeating high-demand SKUs across zones improves workload balance in exactly this profile.
  • Service window pressure. Batch picking still beats zones when order counts are low and SKU diversity is high, because there is no merge point to bottleneck. Wave picking, which groups orders by shipping deadline rather than location, layers on top of either method when you have hard carrier cutoffs to hit.
  • Handoff tolerance. Zone picking’s biggest cost is coordination. Every handoff between zones is a chance for one picker to sit idle while another catches up. If your operation cannot tolerate that synchronization risk, discrete picking, slow as it is, at least fails predictably.
  • The rule of thumb: layer batch picking on top of zones only when a single zone’s SKU velocity is low enough that one picker gathering multiple orders per pass beats waiting for the next tote. Wave logic layers on top of either when deadlines, not SKU location, are your binding constraint.

    Which Metrics Reveal Zone Picking Bottlenecks?

    Four terms show up in nearly every academic treatment of zone picking, and most warehouse managers only track one of them: throughput. That is the gap costing you money.

    Throughput measures orders or lines completed per hour across the whole system. Makespan is the total time to complete a defined batch of orders, start to finish. Blocking happens when a zone finishes its work on a tote but cannot pass it forward because the downstream zone or conveyor slot is occupied. Starving is the mirror image: a zone sits idle because no work has arrived yet. Picker utilization tracks the percentage of a shift a picker spends actively picking versus waiting. Conveyor merge count tracks how often totes converge at a consolidation point, which is often where slowdowns actually originate.

    Research on sequential zone order picking systems found that decomposing workloads into time slots and applying batching and sequencing within those slots measurably reduces blocking and starving delays compared to systems that only balance average workload across zones.

    You can detect blocking and starving without new software. Pull WMS timestamps for tote arrival and departure at each zone, then calculate idle gaps. A short time study, two people with stopwatches for a single shift, will surface obvious imbalance faster than most managers expect. At minimum, instrument these five: throughput per zone, blocking event count, starving event count, picker utilization, and merge count at the busiest consolidation point.

    What Tactical Changes Improve Zone Picking Right Now?

    Most of the gains available to you this month do not require new hardware. They require adjusting five levers that already exist in your operation.

  • Adaptive routing: route totes toward less-loaded zones instead of the shortest path, even when that means a SKU is picked from a zone with lower velocity. This trades a marginal increase in zone visits for a large drop in blocking, and it is the core mechanism behind the adaptive order routing heuristics validated in recent throughput research.
  • Time-slot batching: break your shift into fixed windows and sequence batches within each window rather than releasing work continuously. Temporal imbalance, not average workload imbalance, is what actually produces blocking in most sequential systems.
  • SKU repetition slotting: stock your highest-velocity SKUs in more than one zone so no single zone becomes a chokepoint, and keep fast movers physically close to pack and merge points.
  • Zone-level replenishment thresholds: trigger replenishment when a zone’s aggregate stock drops below a set level rather than watching individual bins, a tactic documented in modern WMS platforms as effective for volatile SKU counts.
  • Merge smoothing: add small buffers ahead of consolidation points, pace tote release intervals, and watch merge counts daily. Reducing unnecessary zone visits, even at the cost of an occasional extra pick, often cuts merge congestion enough to raise net throughput.
  • Pro Tip: Run your first pilot on a single product family for one week before touching anything else. A narrow scope with a clear rollback trigger, like blocking events exceeding a set threshold two shifts in a row, tells you more than a facility-wide change you can’t easily undo.

    Practical guides on picking optimization consistently point back to the same fundamentals: slotting adjustments and pick-path tuning produce measurable gains without capital investment, and they are the right place to start before layering on anything more complex.

    Why Does Orchestration Software Change the Math?

    A warehouse management system tells pickers where inventory sits. It rarely tells them, in real time, which zone is about to starve or which lane is about to block. That gap is what an orchestration layer closes: it sits on top of your existing WMS and makes dynamic routing decisions the WMS was never built to make.

    The evidence here is not theoretical. Adaptive order routing heuristics, tested against common baseline routing in published research, improved throughput by up to roughly 30% while also cutting the number of conveyor merge events. That is a meaningful gap between what static routing rules deliver and what workload-aware, dynamic routing can deliver on the same physical layout.

    Reports suggest efficiency improvements and labor cost reductions are possible for warehouses using orchestration platforms that automate task assignment and enable real-time workforce deployment rather than relying on physical automation.

    An orchestration layer’s practical job is threefold: dynamic order routing that reacts to current zone load instead of a fixed path, workload-aware release timing so totes enter the system only when downstream capacity allows, and cross-zone coordination that a standalone WMS was not designed to handle.

    Before you pilot any orchestration layer, confirm you have:

  • Real-time inventory and location feeds from your WMS.
  • Timestamped event logs for tote movement between zones.
  • A defined pilot window of at least two full shift cycles.
  • A single accountable owner for the go/no-go decision.
  • How Do You Pilot This Without Betting the Whole Operation?

    A pilot only tells you something useful when it is scoped tight and measured consistently. Skip the baseline step and you will not know whether your changes worked or whether volume simply dropped that week.

  • Baseline first. Run a short time study across your target zone or product family and record cycle times, blocking events, and current throughput before changing anything.
  • Scope the pilot narrowly. Pick one product family or one temperature band, apply your new routing or batching rules, and hold everything else constant.
  • Measure over a fixed window. Two full shift cycles is typically enough to separate a real trend from ordinary shift-to-shift noise.
  • Score against four criteria. Track the change in blocking events, throughput per hour, labor cost per order, and any pick-quality metric you already report on.
  • For a rough ROI heuristic: multiply your orders-per-hour improvement by your average labor cost per hour, divide by orders processed, and compare the result to whatever the pilot or software cost you to run. If a modest gain in orders processed per hour covers your pilot cost inside a single quarter, you have a defensible case to expand it facility-wide.

    When Should You Fix the Process and When Should You Adopt a Platform?

    Slotting adjustments and small batching tweaks are often enough when your problem is spatial, meaning one zone is simply overloaded relative to another and a layout or SKU reassignment fixes it. Orchestration software becomes necessary when your problem is temporal: imbalance that appears and disappears within a shift, driven by order timing rather than fixed zone assignments. Static fixes cannot chase a moving target.

    The pitfalls I see most often: managers measure only average workload and miss the short-window spikes that actually cause blocking, pilots get scoped across an entire facility instead of one product family, and conveyor merge behavior gets ignored until it becomes the binding constraint anyway. Fix the measurement problem before you fix anything else.

    — Michael

    Ready to Move Past Manual Zone Rebalancing?

    Slotting and batching fixes buy you real headroom, but they top out once your order mix starts shifting hour to hour instead of week to week. Lully is built for exactly that gap: it plugs into your existing WMS and makes the routing and release decisions a static rule set cannot, rebalancing workload across zones as conditions change rather than waiting for the next manual review.

    If your merge counts and blocking events are already climbing and slotting alone hasn’t closed the gap, a joint time study is the fastest way to find out where orchestration would help most. You can also review operational tactics for conveyor and throughput management as a complementary angle on physical layout while you evaluate software. When you are ready to see what workload-aware routing looks like on your own floor, request a baseline analysis from Lully and get a clear read on where your zones are losing time before you commit to a full rollout.

    Sources

  • Order routing in sequential zone picking systems
  • Optimizing blocking and starving delays in sequential zone order picking systems through time-decomposed workload balancing
  • Warehouse Picking Optimization: Strategies, Tips, and Advice | Exotec
  • Warehousing and storage: NAICS 493 (Bureau of Labor Statistics)