How do you fit 20% more into the same truck? What the three-dimensional packing literature (3D-BPP) gives SMBs: utilization, balance, and delivery-order optimization.
In plain words
Sound familiar?
- Boxes are tossed in at the back of the truck first; the last items don't fit, and some are pushed to the next run.
- The driver plans the load himself; while he's experienced it works, but when he takes leave, utilization drops.
- Customer A's boxes belong deep inside and customer B's at the front, but they get mixed; everything has to be unloaded at every drop.
- Heavy pallets end up on top and fragile boxes underneath; damage occurs and insurance claims rise.
- A truck's utilization rate report says 85% in the spreadsheet; in reality it is below 60-70%, but no one measures it.
- Training a new driver or operator takes 3-6 months; when they leave, the knowledge leaves with them.
- You pay the carrier for the container, and how much product you fit inside directly determines whether you need to order another container.
Why it matters
How it's solved
Technical depth
How it's solved
Technical depthOne-liner: Load heavy, large pallets first and at the bottom; keep fragile and light boxes on top. Group boxes by drop point so the last stop is at the back, the first stop near the door (last-in-first-out). Delivery sequence drives the load order, not raw utilization.
This problem is known in the operations research (a discipline that uses math and computing to solve business-decision problems) literature as the Three-Dimensional Bin Packing Problem (3D-BPP), or in its practical variant the Container Loading Problem (CLP), and it is NP-hard (at large scale a true optimum is unreachable in reasonable time — intelligent shortcuts are essential). The solution is three-stage:
1. Modeling. Each box’s or pallet’s dimensions (width, length, height), weight, orientation constraints (face-down only, no flipping), stacking rules (which boxes can be placed on which), fragility level, delivery order and grouping (multi-drop), the vehicle’s internal dimensions, door height, floor load capacity, and balance constraints (front-back, left-right) are collected into a single data model. Each box is an “item”, each truck or container is a “bin”.
2. Placement via solver. Small scale (50-200 boxes, single vehicle) — a MIP or CP formulation gives near-optimal placements. Medium-large scale (500+ boxes, multiple vehicles) — heuristics and meta-heuristics: extreme-point placement, GRASP, tabu search, simulated annealing. The industry standard is “wall-building” and “layer-building” heuristics combined with a meta-heuristic. The output: which box goes at which coordinate, in which orientation.
3. Field integration. The output is not just a list, it’s a load plan — 3D visualization or layer-by-layer drawings for the driver and warehouse operator, loading order (which box first), weight-distribution report (per axle), and per-stop segmentation. Integration with WMS (warehouse management) and TMS (transportation management) should feed the output to the operator via tablet or printer."
Alternatives
Manual + spreadsheet + driver experience
FreeFree
Who it fits: Small operation loading 1-5 vehicles per day
- + Zero software cost
- + Flexible, accommodates last-minute changes
- + With experienced staff, results are decent
- − Degrades rapidly at scale
- − Knowledge lives in the individual; lost on turnover
- − Inconsistent utilization rate
- − High damage and recall risk
Local WMS/TMS loading module
Enterprise20,000–80,000 TRY license + 4,000–10,000 TRY/year maintenance (TR market observation)
Who it fits: Medium scale (10-50 vehicles daily, limited variety of box sizes)
- + Local support, native team
- + Integrates with the WMS
- + Heuristic-based, fast results
- − Weak on complex stacking rules
- − 3D visualization usually limited
- − Multi-drop delivery sequencing support is variable
International specialized container-loading software
Enterprise15,000–80,000 EUR license + 5,000–15,000 EUR/year
Who it fits: Large fleet, exporting SMB, 3PL operator
- + Mature 3D engine, weight distribution, damage rules
- + Multi-mode: container, truck, rail car
- + Full load plan and visualization
- − Expensive
- − 3-6 month deployment
- − Local-language support usually weak
Open-source solvers + custom application
Open SourceLicense free; in-house build 12-20 weeks or 300K-900K TRY consulting
Who it fits: SMB with a technology team and many custom rules
- + No license, unlimited customization
- + Business rules in code, changeable on demand
- + Mature open-source MIP solvers and specialized heuristics available
- − In-house OR/software capacity required
- − 3D visualization UI built separately
- − Ongoing maintenance burden
Recommendation
Ask in the meeting
- How does the software handle box orientation constraints (this-side-up, no flipping)? Is it visible in the visualization?
- What level of weight-distribution control? Do you report front and rear axle limits separately?
- Multi-drop delivery: for a load with 10 unload stops, does the loading sequence get optimized automatically?
- Can custom rule sets like fragile / no-stack / cold-chain be defined separately?
- WMS / TMS / ERP integration via which protocols? Do you have an API or is it file-based?
- Can I switch the optimization target: today maximum utilization, tomorrow minimum damage, the next day minimum loading time?
- Can we run a 2-4 week pilot with real data during the trial period? How are success criteria defined?
- If we stop working with you, how do we get our box master data and historical load plans back? Is there a standard format export?
Technical details
Editor’s note
Colloquially this problem is called “truck loading plan”, “container filling”, or “pallet layout”. In the academic literature it is the Three-Dimensional Bin Packing Problem (3D-BPP) or, with practical side constraints, the Container Loading Problem (CLP). Without knowing both names share the same underlying math, you can’t test whether the “superior algorithm” a vendor pitches actually uses sound heuristics and handles the side constraints you need.
The most-skipped point in the industry: the weight-distribution (balance) constraint. A truck can be 95% full by volume yet illegal to drive because it exceeds an axle weight. Software that maximizes volume utilization without checking center of gravity is, in reality, doing half the job.
Step-by-step path — for the SMB
Stage 1 — Measure first, plan after. Log every vehicle loading for at least 4 weeks: vehicle type, exit utilization rate (volume and weight), loading time, damage incidents, delivery-sequence complaints. Current knowledge assets: which box can go on which, whose delivery sequence matters, which vehicle dimensions differ.
Stage 2 — Extract the knowledge asset. Box master data (width × length × height × weight), orientation, stacking rule, fragility level, customer group. Without this data, no software produces good output. The vendor will ask for it on day one.
Stage 3 — Pilot. 8-12 weeks. One depot, one route type first. Success criteria defined in advance: utilization +5% minimum, loading time -20% minimum, damage rate preserved or lower. A driver and a warehouse operator are selected as “champions”.
Stage 4 — Scale-out. 3-6 months to full warehouse and fleet. The 3D load plan reaches the operator via tablet or screen. Data flows are closed with ERP, WMS, and TMS.
Risks — what can go wrong
- Data quality. Box dimensions should be measured, not nominal (the supplier’s “30×40×20” vs. actual 32×42×21 produces 5-8 m³ of error on 1000 boxes). All box dimensions should be re-measured in the first week.
- Operator resistance. “I’ve been loading for 30 years, the computer is going to teach me?” is a frequent reaction. During the pilot, experienced operators should be used to evaluate the software output; initially position it as advisory, not mandatory.
- Integration time. Vendor says “we’ll integrate with the WMS in a week”; the reality is 4-8 weeks. The contract should include an explicit integration schedule and test environment.
- Single-vendor lock-in. Box master data and historical load plans are critical knowledge assets. The contract should include “standard format export rights, customer data belongs to the customer”.
Solution method — a technical look
| Approach | Typical scale | Solve time | Guaranteed optimum? |
|---|---|---|---|
| Exact MIP / CP | 50-200 boxes, single vehicle | minutes-hours | Yes (within limit) |
| Heuristic (wall-building, layer) | 200-2000 boxes | seconds-minutes | No, 85-95% optimum |
| Meta-heuristic (GRASP, tabu, SA) | 500-10000 boxes, multi-vehicle | minutes | No, 90-97% optimum |
| Heuristic + 3D viz combined | Industry-common solution | minutes | No |
Objective function choice:
- Objective 1 — Maximum volume utilization: Move more goods per vehicle. Default for most SMBs.
- Objective 2 — Minimum vehicle count: Pack the total load into the fewest vehicles. Cost-minimization focus.
- Objective 3 — Minimum damage / stacking violations: Load without breaking fragility or weight-bearing rules. Insurance and customer-complaint focus.
- Objective 4 — Multi-drop sequencing: Last drop deepest, first drop closest to the door. Delivery-time focus.
Good software makes the objective function configurable; you can re-weight every day, every customer, every vehicle type.
Academic references
See the sources field in the frontmatter.
Sources
- Bischoff, E. E. and Ratcliff, M. S. W. (1995). Issues in the development of approaches to container loading. Omega, 23(4), 377–390. Foundational reference for the practical container-loading literature.
- Martello, S., Pisinger, D. and Vigo, D. (2000). The three-dimensional bin packing problem. Operations Research, 48(2), 256–267. Classical 3D-BPP exact approach.
- Wäscher, G., Haußner, H. and Schumann, H. (2007). An improved typology of cutting and packing problems. European Journal of Operational Research, 183(3), 1109–1130. Classification of packing problems.
- Turkish Statistical Institute — Road Freight Transport Statistics (annual). Reference for domestic transport volumes and utilization indicators in TR.
- Turkish CoHE (YÖK) Thesis Center — keywords: ’three-dimensional packing’ or ‘container loading’ — 40+ Turkish masters / doctoral theses. tez.yok.gov.tr
Glossary
- Bin Packing
- The problem of placing objects of various sizes into fixed-capacity bins so as to minimize the number of bins used.
- Container Loading
- The problem of placing parcels and pallets into one or more containers or trucks; the 3D-BPP extended with practical side constraints.
Related problems
From One Node to Another — How Do I Compute the Shortest Path on a Weighted Graph?
For SMBs that need to compute the fastest or shortest route between two points: 10-50-vehicle field-service teams (plumbing, electrical, appliance repair), urban courier/parcel operations, or dispatch centres coordinating emergency response. Every day brings hundreds of 'how do I get from A to B fastest right now' questions; the answer shifts with traffic, road closures and vehicle type. A wrong route costs the technician one or two jobs missed for the day, the courier a late delivery, and the firm a customer. Manual or by-eye routing typically leaves 20-60 wasted minutes per vehicle per day on the table compared with a network-aware route calculation.
Multiple Plants, Multiple Customers — How Much Does Each Plant Ship to Each Customer to Minimise Total Freight?
For food, packaging or textile producers shipping weekly from 3-8 plants or regional warehouses to 20-100 customers. The weekly decision is: which plant ships how much to which customer, given fixed plant capacities, stated customer demands, and a different per-unit cost (distance + vehicle + contract terms) for each plant-customer pair. The goal is the lowest total freight bill across the network. A 'nearest plant' or 'we've always done it this way' habit typically leaves 10-20% extra fuel and vehicle cost on the table compared with a systematic allocation.
One Vehicle, Many Stops — In What Order Should I Visit Them All to Minimize Total Distance?
You run a field technician visiting 8-15 customers a day (HVAC, lift servicing, white-goods repair), a single-vehicle supplier tour by a sales rep, or a PCB drilling machine sequencing 500-5,000 holes. All of them face the same core call: given N points, in what order should a single vehicle or head visit each one and return to the start. Get the order wrong and a field service vehicle burns 80-200 TRY/day extra in fuel and driver hours, a PCB line takes 15-30% longer per part, and the last customer of the day misses their delivery window. At 50 stops, a hand-built sequence runs 20-40% above the true minimum; as the number of stops grows, the gap from intuitive ordering compounds.