Carton Flow Rack: Pick Rates, Lane Sizing and When It Beats Shelving

Ask a warehouse manager where their picking hours actually go and almost nobody says “lifting.” They say walking, hunting for the right label, and waiting behind a replenishment cart. That is the problem carton flow rack was built for. It is not faster shelving; it is a different geometry that removes the two slowest parts of a case pick.

This is a spoke in the same family as pallet flow racking, which does the same trick with full pallets and deep lanes instead of cases. The engineering rules are similar. The sizing rules are not, because a carton is small enough that a half inch of lane clearance, or two wheels too few under the box, decides whether the system flows or stalls.

A static shelf holds inventory. A flow lane presents it. That distinction is the whole value case. On static shelving a picker walks the aisle, scans for a label, reaches into the shelf depth, and often has to shift the front cases to get at the one the system wants. In a flow lane, every SKU has a dedicated front position, the face label never moves, and the next case is already at the stop when the last one leaves.

The practical effect is that a pick module built on flow lanes compresses the same number of SKUs into a much shorter pick face. A 40-foot run of four-tier static shelving might hold 60 case SKUs. The same SKUs in five tiers of eight-foot-deep flow lanes fit in roughly half the linear feet, because the depth is carrying inventory instead of the width. Halving the pick face halves the walking, and walking is where the seconds are.

That density has a second consequence worth planning for. When the pick face shrinks, the replenishment traffic concentrates behind it. Flow rack only pays if the restock aisle is genuinely separate, which is why the layout question comes before the equipment question.

This is the selection decision that the catalogue pages skip, and it is the one that determines whether cases stall mid-lane six months after installation. The choice is driven by the carton — its bottom, its length, its weight and how much that varies across the SKUs you plan to put in the bay.

The governing rule is simple: at least three points of support must be under the case at all times, and the case bottom must be rigid enough to span the gap between them. Everything else follows from that.

Lane sizing is arithmetic, not judgement, and it starts from your actual case dimensions — not the nominal ones on the purchase order. Measure a real case, at the widest point, with tape and any bulge included.

Lane width is carton width plus about one inch of total clearance, half an inch per side, plus the thickness of the divider. Go tighter and a slightly over-packed case jams. Go looser and the case yaws across the lane and hits the divider at an angle, which is the same failure by a different route. Vertical clearance between the lane and the shelf above should be carton height plus roughly two inches so a picker can get a hand on the case without lifting it clear.

Lane depth is a stocking decision. Depth should hold between half a day and two days of demand for that SKU — enough that replenishment is a scheduled task rather than an interruption, not so much that slow movers tie up the most valuable real estate in the building. In practice that lands most lanes between four and twelve feet, and it means your fastest SKUs should get the deepest lanes at the best height, not the most lanes.

Pitch is where most retrofits go wrong, because the installer sets one slope for the whole bay and the bay holds a mix of 3-pound and 30-pound cases. A slope that moves the heavy SKU reliably will let the light SKU stall halfway down; a slope that moves the light SKU will let the heavy one arrive at the stop hard enough to burst. If the weight spread across a bay is wider than roughly ten to one, split the bay and set two pitches, or move the heavy SKUs to a shallower lane where momentum has less room to build.

Nobody on the first page of search results will do this arithmetic for you, so here it is. A case pick decomposes into four times: travel between picks, search and identify, grasp and extract, and place into the carton or pallet. Carton flow attacks the first two hard and the third moderately. It does nothing at all for the fourth.

Take a realistic static-shelving baseline. Average travel between consecutive picks of 18 feet at a working walk of about 200 feet per minute is 5.4 seconds. Scanning a shelf face for the right label and confirming the SKU is about 4 seconds. Reaching into a 24-inch-deep shelf, shifting the case in front and extracting the one you want is about 6 seconds. Placing it is about 4 seconds. Total: roughly 19.4 seconds, or 185 picks per hour at the face.

Now the flow lane. The pick face is half as long, so travel drops to about 6 feet — 1.8 seconds. The label is permanent and at eye level in a known position, so search falls to about 2 seconds. The case is already at the stop in the golden zone, so extraction is about 4 seconds. Placing is unchanged at 4 seconds. Total: roughly 11.8 seconds, or 305 picks per hour.

That is a 39% reduction in seconds per pick. Be careful how you bank it. The model above is the pick face only; a real shift also contains travel between zones, paperwork, equipment moves and breaks, so a face-level gain of 39% typically shows up as a 20% to 30% improvement in measured lines per labour hour. That is still the largest single improvement most conventional pick operations can buy without automation, and it is why carton flow carries the highest cost-per-click of any racking term.

The extraction number is not a guess either — it is ergonomics, and it is quantified. The NIOSH revised lifting equation scores a lift by multiplying a load constant by a set of penalties. The vertical multiplier is highest at a 30-inch origin height and drops by 0.0075 for every inch above or below it, so a case picked off the floor at 6 inches scores 0.82 and a case picked at 60 inches scores 0.78. The horizontal multiplier is 10 divided by the horizontal distance in inches: a case held 10 inches from the body scores 1.00, while the same case reached 25 inches into a deep shelf scores 0.40.

That second number is the entire argument for gravity feed. Reaching into shelf depth is not slightly worse than picking from a stop at the lane face — it is a 60% penalty on the safe lifting limit for that task. Presenting the case at the front of a lane, between roughly 24 and 66 inches, converts a deep reach into a short one on every pick of every shift.

A flow lane has two sides and they belong to two different jobs. That separation is the second half of the productivity gain, and the half that layout mistakes usually destroy.

Restocking a static shelf means standing in the pick aisle with a pallet jack, blocking the aisle for everyone behind you, and often pulling the front stock forward to get the new stock behind it. Restocking a flow lane means walking the charge aisle behind the bay and pushing cases in. Nobody stops picking. First in, first out is enforced by the geometry rather than by discipline, which matters for anything date-coded.

Most carton flow in California is not a new building — it is flow shelves dropped into pallet rack bays that are already standing. That is the cheapest way to buy the pick rate, and it carries two obligations that a new installation would have settled at the design stage. Carton flow suits cartons and totes; when the load is pipe, lumber or sheet stock the geometry changes completely, and cantilever arm capacity sizing is the calculation that governs the rack instead.

The first is structural. Flow lanes change how load sits in the bay: instead of two pallets landing on the beams at known points, you have a distributed live load across the full shelf depth, plus the weight of the track itself. Cal/OSHA Title 8, Section 3241 requires racks to carry the manufacturer’s rated load posted on the rack and prohibits exceeding it, so the retrofit has to be signed off against the frame and beam capacity you actually own, not the capacity of the rack you think you own. Get the frame series, beam size and current capacity plate before anyone quotes a shelf.

The second is permitting. In most California jurisdictions, changing a racking configuration is a plan-check matter and seismic bracing is part of the calculation. Settle the warehouse racking permits question before the steel is on the truck; it is a scheduling problem, not usually a blocking one, but it is a four-week scheduling problem if you discover it late.

The two systems answer different questions and are routinely confused in quoting. Carton flow is a case-picking device: cases picked individually off a front stop, lanes four to twelve feet deep, one SKU per lane, sized in inches. Pallet flow racking is a storage and staging device: full pallets moving on heavy rollers with speed controllers, lanes often two to twenty pallets deep, sized in pallet positions. Carton flow earns its keep on broken-case picking; for full-pallet storage of a handful of high-volume SKUs, Drive-In Racking buys considerably more density per square foot.

A high-throughput operation frequently runs both in the same building — pallet flow feeding reserve and staging at the dock, carton flow at the pick module — and the boundary between them is the point where you stop moving a pallet and start moving a case. Draw that boundary from your order profile: the percentage of order lines that ship as full pallets versus as cases is the number that decides how much of each you need.