Skip to main content

Grain Silo Packing

BIN PACKING PROBLEM

Efficient silo utilization reduces storage costs by 10–20% at commercial grain elevators handling 50,000–200,000 bushels across 10–30 bins (Maier & Bakker-Arkema, 2002). Every harvest requires deciding how to assign dozens of grain lots—each a different crop type and tonnage—into a limited number of silos without mixing incompatible grains. This is the 1D Bin Packing Problem (BPP)—one of the foundational strongly NP-hard problems in combinatorial optimization.

Where This Decision Fits

Post-harvest operations chain — the highlighted step is what this page optimizes

Harvest Field collection
Grain Drying Moisture reduction
Quality Grading Test weight & grade
Silo Assignment Pack lots into bins
Storage Monitoring Temperature & aeration
Market & Shipping Sale & transport

The Problem

From grain silos to optimization theory

A grain storage facility has silos each holding 50 tons. After harvest, 12 grain lots of varying sizes must be stored. The goal is to minimize the number of silos used while fitting all lots.

This is the 1D Bin Packing Problem: pack n items into bins of capacity C using the minimum number of bins. Strongly NP-hard.

Agriculture DomainBin Packing Model
Grain silo (50t)Bin (capacity C)
Grain lotItem (size si)
Minimize silos rentedMinimize number of bins
BPP — Strongly NP-hard; FFD gives 11/9 · OPT + 6/9 approximation

Try It Yourself

Edit grain lot weights, adjust silo capacity, and find the optimal packing

Grain Lots & Silo Configuration

12 Lots · Click any cell to edit
Choose a Scenario
A typical post-harvest intake day at a grain elevator: 12 lots of varying size across corn, wheat, soybean, and canola. Silo capacity is 50 tons, and the goal is to minimize silos used.
Silo Capacity: 50 tons
LotCropWeight (tons)
Select Algorithm
Silo Allocation

The Algorithm

First Fit Decreasing (FFD) for Bin Packing

FF (First Fit) Items placed in arrival order — gaps accumulate Silo 1 Silo 2 Silo 3 Silo 4 18t 12t 8t 25t 15t 30t empty 3 silos · gaps remain FFD (First Fit Decreasing) Sorted by size first — large items placed early Silo 1 Silo 2 Silo 3 30t 18t 25t 15t 8t 12t 3 silos · 96% utilization Same 6 lots (108t total) — FFD packs tighter by placing large items first
1

Sort Items by Size

Arrange grain lots in decreasing order of weight. Large lots are placed first.

2

First Fit Placement

For each lot, scan silos from first to last. Place the lot in the first silo with enough remaining capacity.

3

Open New Silo if Needed

If no existing silo can fit the lot, open a new silo and place it there.

4

Return Assignment

The resulting assignment uses at most 11/9 · OPT + 6/9 bins. Very effective in practice.

Real-World Complexity

Factors beyond the basic bin packing model

Grain Compatibility

Some grains cannot be mixed in the same silo due to contamination risk.

Temperature Control

Different grains need different storage temperatures and humidity levels.

Turnover Schedule

Silos are emptied at different times; packing changes dynamically.

Pest Control

Silo fumigation schedules create temporary unavailability.

Silo Sizes Vary

Real facilities have silos of different capacities, not just one type.

Loading/Unloading

Silo access points limit how quickly grain can be moved in and out.

References

Key literature on bin packing

Johnson, D.S. (1973).
"Near-optimal bin packing algorithms."
PhD Thesis, MIT.
Coffman, E.G., Garey, M.R. & Johnson, D.S. (2013).
"Bin packing approximation algorithms: survey and classification."
Handbook of Combinatorial Optimization, Springer.

Need to optimize grain
storage allocation?

Get in Touch
Data shown is illustrative. This is a simplified model for educational purposes.
 Home
ESC