VIEW THIS AS

Auto mode follows the Route Engine until you choose a viewpoint.

YOU ARE HERE

ROUTE CHECK

CONNECTED TO

WHAT NEXT

Use the canonical route for this room, or HELP if you are unsure.

Why Mathematics? | Origami, Crease Patterns and Fold Geometry

Three students in school uniforms work through open books at a classroom table, with textbooks and stationery nearby and study notes on the whiteboard behind them.

Why is mathematics important in origami? A folded paper form may look artistic and intuitive, but every crease carries geometric information. A fold can reflect a point to another point, create a perpendicular bisector, divide an angle, constrain several panels to rotate together, or turn a flat sheet into a structure that expands and contracts.

Origami therefore makes mathematics physical. Lines become hinges, regions become faces, intersections become vertices, and an apparently simple instruction—“bring this point to that line”—can encode an equation. The same ideas support paper art, deployable structures, folded products, robotics and research into thin-sheet mechanics.

This article builds the subject from first principles. We will explore reflections, crease constructions, flat-foldability conditions, graphs, rigid motion, thickness and optimisation. We will also separate ideal mathematical paper from real material, because an elegant theorem does not cancel collisions, tearing or thickness. The happiest part is that students can investigate many ideas with a square of paper, a ruler, careful diagrams and a willingness to test a claim.


Choose the fold question you want to solve

Origami mathematics contains several related but distinct questions.

  • If you want to construct a midpoint or perpendicular bisector, begin with folds as reflections.
  • If you want to divide an angle, examine how one line can map onto another.
  • If you want a crease pattern to lie flat, study sector angles and mountain–valley assignments.
  • If you want a folded form to move without bending its panels, study rigid origami and degrees of freedom.
  • If you want to repeat a pattern, use symmetry, transformations and tessellation.
  • If you want to design a usable object, include paper thickness, material strain, collisions and manufacturing tolerances.
  • If you want to discover a pattern experimentally, follow the student project near the end.

A single design can involve all of these. Keeping the questions separate makes the mathematics clearer: constructibility, flat-foldability, rigid motion and practical manufacturability are not synonyms.


An ideal fold acts like a reflection

Imagine folding point A exactly onto point B. Every point on the crease is equally distant from A and B, so the crease is the perpendicular bisector of segment AB.

In coordinates, let A = (2, 1) and B = (8, 5). Their midpoint is:

M = ((2+8)/2, (1+5)/2) = (5, 3).

The slope of AB is:

(5−1)/(8−2) = 4/6 = 2/3.

A perpendicular line has slope −3/2, so the crease through M is:

y − 3 = −(3/2)(x − 5).

The physical action of matching two points has constructed a line that algebra could also describe. This is why paper folding can be a serious geometry laboratory: a motion embodies a transformation.


Equal distances explain the perpendicular bisector

Take any point P on the crease. Folding reflects A to B while leaving P fixed because P lies on the hinge. Reflection preserves distance, so PA = PB.

Conversely, the set of points equidistant from A and B is their perpendicular bisector. This “locus” viewpoint is important. A crease is not merely the visible mark left by paper; it is the solution set of a condition.

Students can test the claim by selecting several points along the crease and measuring each distance pair. Measurement will not be exact because paper and rulers have limited precision, but the differences should cluster near zero. The discrepancy becomes evidence about experimental error rather than a reason to abandon the theorem.


Folding one line onto another creates an angle bisector

When a fold maps line l₁ onto line l₂, the crease is positioned so that reflection across it sends one line to the other. For intersecting lines, the possible crease lines are their angle bisectors.

There are normally two bisectors: one for the smaller pair of vertical angles and one for the larger pair. They are perpendicular to one another. A paper construction may reveal both possibilities depending on the direction of the fold.

If the angle between the original lines is 74°, the internal bisector creates two 37° angles. Yet the important reasoning is not “divide 74 by two” alone. Reflection makes corresponding rays coincide, so their angles to the crease must be equal.


Folding a point onto a line creates a family of possible creases

Suppose point P must land somewhere on line l. The destination is not fixed, so many folds may satisfy the condition. Each possible crease is the perpendicular bisector of P and its chosen image point on l.

The family of such creases is related to tangents of a parabola whose focus is P and whose directrix is l. For a point X on a crease, equal-distance reasoning connects X to P and to the corresponding point on the directrix.

This is a beautiful bridge from elementary folding to conic sections. A repeated sequence of crease lines can envelope the shape of a parabola. The paper does not draw the curve directly; the curve emerges as the boundary touched by a family of straight lines.


Some origami constructions solve equations beyond straightedge and compass

Classical straightedge-and-compass constructions have well-known algebraic limits. Certain paper-folding operations impose several conditions at once—for example, placing one point onto one line while placing another point onto a second line. The resulting crease can correspond to solving a cubic equation.

This does not mean any casual fold magically solves every cubic. The construction requires the correct geometric setup, may have multiple valid folds, and inherits physical error. It does show that a different allowed operation changes what is constructible.

The broader mathematical lesson is powerful: the rules of a system determine its capabilities. Change the permitted operations, and the set of solvable problems may change too.


A crease pattern is a geometric network

Before folding, a crease pattern can be represented as a planar arrangement of vertices, edges and faces.

  • A vertex is an intersection or endpoint of creases.
  • An edge is a crease segment between vertices or boundaries.
  • A face is a region of paper bounded by edges.
  • Boundary segments also matter because the sheet is finite.

This graph-like description separates topology—what connects to what—from exact geometry—lengths and angles. Two patterns can share the same connectivity while using different sector angles. They may therefore behave differently when folded.

For a connected planar graph drawn on a sheet and treated with the appropriate exterior face, Euler’s relationship V − E + F = 2 can provide a counting check. Boundary conventions matter, so students should state exactly what they count rather than applying the formula mechanically.


Sector angles describe the space around a vertex

At an interior vertex, creases divide the full 360° turn into sector angles. If there are four creases, the sectors might be 60°, 90°, 120° and 90°.

The first basic check is:

60 + 90 + 120 + 90 = 360°.

That confirms the sectors fill a complete turn. It does not yet prove the vertex can fold flat. Flat-foldability adds an alternating-angle condition.

This is a useful modelling pattern: begin with a universal geometric constraint, then add conditions for the specific behaviour you want.


Kawasaki’s theorem tests alternating sector angles

For a single flat-foldable interior vertex in ideal zero-thickness paper, Kawasaki’s theorem says the alternating sector angles sum to 180°.

Using the example 60°, 90°, 120°, 90° in circular order:

60° + 120° = 180°

and

90° + 90° = 180°.

So the alternating-angle condition is satisfied.

For a single vertex, this condition is a remarkably efficient geometric test. In a larger crease pattern, however, locally valid vertices may still conflict globally through layer order, boundary constraints, panel collisions or incompatible motions. Local success is necessary information, not a complete guarantee for an entire object.


Maekawa’s theorem constrains mountains and valleys

Creases are commonly labelled mountain folds or valley folds. At a flat-foldable interior vertex, Maekawa’s theorem states that the number of mountain creases and valley creases differs by two:

M − V = ±2.

For a degree-four vertex, a valid count can be three mountains and one valley, giving 3 − 1 = 2, or the reverse.

The theorem constrains the count, not the entire assignment. If a vertex has many creases, several mountain–valley patterns may satisfy the count while only some work with the sector geometry and neighbouring vertices. Combinatorics and geometry must cooperate.


Flat-foldability is not the same as rigid foldability

A pattern is flat-foldable if it can reach a flat state under the mathematical model. A pattern is rigidly foldable if its faces can move like rigid panels connected by hinges, without bending the faces.

Some paper models reach a flat configuration by flexing panels slightly. That behaviour may be acceptable for craft paper but unsuitable for a mechanism made from metal or composite panels. Conversely, a rigidly moving pattern may have a limited range before collision and may not collapse fully flat.

Research on rigid origami studies the compatibility of rotations about creases and the configuration space of a crease pattern. The distinction is essential for engineering: a theorem about the final flat state does not automatically describe the motion used to reach it.


Degrees of freedom count independent motion parameters

A free rigid body in three-dimensional space has translations and rotations. Hinges and connections remove possible motions by adding constraints. In a folded mechanism, the remaining independent parameters are its degrees of freedom.

A simple book-like hinge has one obvious degree of freedom: the opening angle. A multi-crease vertex may also move with one coordinated parameter, meaning that changing one fold angle determines the others through compatibility equations.

Counting variables minus independent constraints can suggest the number of degrees of freedom, but dependencies and special geometries can make naive subtraction fail. Engineers therefore use kinematic analysis, Jacobian matrices and numerical continuation for complicated systems.

For students, the practical question is accessible: if you hold one panel fixed, how many independent choices are needed to determine the whole configuration?


Fold angles are three-dimensional variables

A crease pattern drawn flat contains sector angles in the plane. Once folded, each crease also has a dihedral or fold angle describing the relative orientation of adjacent faces.

Mountain and valley labels can be represented by signs under a chosen convention. The exact numerical sign convention varies, so a report must define it. A fold angle of zero may represent the unfolded state in one model, while another convention uses 180°.

This illustrates a general lesson about mathematical notation: symbols become meaningful only after definitions. Two programs can describe the same physical fold using different angle origins and signs. Comparing their raw numbers without translating conventions would be a mistake.


Rotations around a vertex must be compatible

Moving from one panel to the next across a crease applies a rotation. After travelling around all creases incident to an interior vertex, the sequence of rotations must return consistently to the starting panel.

In matrix language, an ordered product of rotation matrices must satisfy a closure condition. Multiplication order matters because three-dimensional rotations do not generally commute. Rotating around one axis and then another can produce a different orientation from reversing the order.

This is one reason rigid-origami mathematics can become sophisticated even when the paper looks simple. Local angles are linked by nonlinear compatibility equations. The physical model gives a tangible route into transformations and matrices.


Symmetry reduces a complicated pattern

Many origami designs repeat a module through translation, rotation or reflection. If identical units move symmetrically, analysis can focus on one representative cell plus boundary conditions.

For example, a periodic tessellation might repeat every a units horizontally and b units vertically. Translational symmetry means coordinates differ by integer multiples of vectors (a,0) and (0,b). A rotationally symmetric vertex may repeat sector and fold-angle relationships.

Symmetry can reduce variables, simplify manufacturing and make motion easier to understand. But assuming symmetry too early can hide asymmetric modes or defects. It is a modelling choice that should be tested, not merely admired.


Tessellations connect local rules to global patterns

An origami tessellation repeats folds across a sheet. The local unit must meet neighbours without gaps or overlaps in the flat pattern, and its folded motion must be compatible across shared edges.

This creates a local-to-global challenge. A vertex can satisfy Kawasaki’s and Maekawa’s conditions, yet repeating it may create layer-order conflicts or lock the structure. Conversely, a carefully designed unit can generate coordinated contraction across a large sheet.

Students can explore this by designing one quadrilateral unit, copying it by translation and checking every shared edge. The exercise links transformations, congruence, periodicity and error accumulation.


The Miura pattern shows coordinated expansion and contraction

The Miura-ori is a well-known repeating arrangement of parallelogram facets. Under an ideal rigid model, its panels can fold in a coordinated motion, allowing a sheet to compact in one direction while changing in the other.

Researchers study such patterns for deployable structures and unusual mechanical responses. The exact relationship between dimensions and fold angles depends on the pattern parameters and modelling assumptions.

It is tempting to say “origami makes every structure stronger” or “the Miura fold solves deployment”. Those claims are too broad. Performance depends on material, thickness, loads, joints, boundary conditions, scale and manufacturing. The mathematics helps engineers evaluate a design; it does not guarantee success from the pattern name alone.


Area stays constant in ideal folding, but projected area changes

Ideal folding does not stretch or cut the sheet, so its surface area remains constant. Yet the area of its shadow or projection onto a plane can shrink dramatically.

Consider a rectangular sheet of area A = LW. When panels tilt out of the plane, a panel with area Aᵢ and angle θᵢ to the target plane contributes a projected area involving Aᵢ cos θᵢ, provided overlaps are handled separately.

When panels overlap in projection, simply summing projected areas double-counts shared regions. Geometry therefore distinguishes material area from footprint, visible area and enclosed volume. A compact folded object can contain the same sheet area while occupying a much smaller plan view.


Folding in half produces exponential layer growth

In an idealised thought experiment, folding a sheet exactly in half doubles the number of layers. After n folds, the layer count is:

L = 2ⁿ.

After ten folds, the model gives 2¹⁰ = 1024 layers. If each layer has thickness t, a naive stack thickness would be 2ⁿt.

Real paper rapidly resists this model. Bending radii, trapped air, misalignment, compression, finite sheet dimensions and required force become important. The exponential formula describes layer counting under an ideal rule; it does not prove that unlimited folding is physically possible.

This makes a fine lesson in model boundaries. A correct sequence can still become physically infeasible.


Real thickness changes crease placement

Mathematical origami often treats paper as having zero thickness. Actual materials have thickness, and multiple layers accumulate. A fold that aligns perfectly in the ideal model may shift because outer layers travel farther around a bend than inner layers.

Thick-origami design may introduce hinge gaps, offsets, tapered panels or modified crease locations. The correction depends on thickness, fold angle and stacking order. Flexible paper can absorb small inconsistencies through deformation, but rigid panels cannot.

This is not a failure of geometry. It is a reminder that a useful physical model may need additional variables. Thickness converts an abstract line into a finite hinge region.


Materials bend, stretch, crease and fatigue

An ideal face remains perfectly rigid while an ideal crease rotates freely. Real sheets distribute curvature over a small region, and repeated folding can weaken fibres or create plastic deformation.

Designers may need to model bending energy, stretching energy, yield limits and fatigue. Thin-sheet mechanics often involves a competition: bending can be relatively easy, while stretching the sheet is much more costly. This difference helps explain why paper forms sharp creases and develops buckling patterns.

A classroom paper model is therefore evidence about geometry and qualitative motion, not a full structural test. Claims about loads, medical devices, vehicles or space hardware require the relevant material and safety validation.


Layer order is a hidden combinatorial problem

When a model folds flat, multiple faces occupy overlapping regions. The geometry must decide which layer lies above another. Pairwise ordering constraints can be represented as a directed graph.

If face A must lie above B, B above C, and C above A in the same region, the cycle signals an impossible ordering. Even without such a simple cycle, different overlapping regions can make global consistency difficult.

This is why a crease pattern that passes local angle tests may still fail as a physical flat fold. Layer order adds discrete logic to continuous geometry.


Collision detection asks whether panels occupy the same space

During folding, two non-adjacent faces may intersect even if the start and end configurations look valid. A motion planner must test intermediate configurations.

Computational approaches represent panels as polygons in three dimensions and check for intersections as fold parameters change. Finer time steps reduce the chance of missing a brief collision but increase computation. Continuous collision methods attempt to reason over an interval.

The same problem appears in robotics and computer graphics. Origami offers a compact setting in which geometry, algorithms and physical constraints meet.


Inverse design starts with the desired behaviour

Forward design asks, “What shape or motion will this crease pattern produce?” Inverse design asks, “What crease pattern will produce this target shape, footprint or movement?”

The inverse problem can have many solutions or none under the chosen constraints. An objective function might penalise distance from a target surface, excessive strain, too many unique parts or poor compactness. Constraints might enforce flat-foldability, rigid panels, minimum feature sizes and non-collision.

Optimisation then searches a large design space. A low objective value is not enough; the candidate must be checked for manufacturability and sensitivity. The “best” design depends on how the objective and constraints were defined.


Fold errors can accumulate across a pattern

Suppose each repeated crease is misplaced by a small random amount. Across many cells, some errors may cancel while others accumulate. Systematic bias—such as every crease shifted in the same direction—can be more damaging than zero-mean random noise.

Students can quantify this by folding repeated strips, measuring endpoint displacement and plotting error against the number of units. Repeating the experiment estimates variability. A regression may suggest a trend, but the conclusion should distinguish systematic drift from random scatter.

This is a meaningful bridge from geometry to statistics. A theoretically perfect pattern becomes a manufacturing process with tolerances.


Origami applications need evidence, not magic words

Researchers and engineers study folding concepts for deployable arrays, compact structures, soft robots, medical devices, packaging and architecture. These applications do not all use the same material or mathematical model.

A space structure may prioritise compact stowage, reliable deployment and thermal behaviour. A medical device may prioritise biocompatibility, fatigue, precise actuation and clinical validation. A carton may prioritise material use, automated folding and protection. A robot may require controllable motion and sensors.

Calling all of them “origami-inspired” identifies a design influence. It does not establish performance. Responsible writing names the exact mechanism and cites the relevant testing.


Common misconceptions

“Origami mathematics is only about decorative symmetry”

Symmetry is important, but folding also involves reflection, constructibility, graph theory, kinematics, optimisation, mechanics and computation.

“If every vertex is locally flat-foldable, the whole pattern must fold flat”

Local conditions do not resolve every global layer-order, boundary or collision constraint.

“Kawasaki’s theorem and Maekawa’s theorem say the same thing”

Kawasaki constrains alternating sector angles. Maekawa constrains the difference between mountain and valley counts.

“Flat-foldable means rigidly foldable”

Flat-foldability concerns reaching a flat state. Rigid foldability concerns a motion with unbending faces. A model may have one property without the other.

“A paper prototype proves an engineering structure will work”

Paper can demonstrate geometry, but scale, thickness, material properties, loads, joints and safety tests remain.

“Zero-thickness paper is an error”

It is an intentional idealisation that makes many theorems tractable. It becomes inappropriate when thickness controls the result.

“More creases always create more flexibility”

Extra creases add possible hinges but also compatibility constraints. They can introduce motion, redundancy or locking depending on the geometry.

“A computer-generated pattern is automatically manufacturable”

Optimisation output still needs minimum feature sizes, tolerances, collision checks and material validation.


A six-week origami mathematics project

Use ordinary paper and safe classroom tools. Do not cut toward hands, force thick materials or claim a paper model is load-bearing.

Week 1: prove a reflection construction

Mark two points, fold one onto the other and trace the crease. Measure several points on the crease to compare their distances from the original points. Explain why the crease is a perpendicular bisector.

Week 2: explore angle bisection

Draw two intersecting lines and fold one onto the other. Measure the resulting angles and distinguish the internal and external bisectors.

Week 3: test a one-vertex pattern

Choose four sector angles adding to 360°. Create one set satisfying Kawasaki’s alternating sums and one that does not. Assign mountains and valleys, then compare observed behaviour with Maekawa’s count.

Week 4: repeat a module

Design a small cell and translate it across a grid. Check shared edges, sector angles and consistency. Record any mismatch caused by construction error.

Week 5: measure motion

Select one fold angle as an input and photograph or measure the changing width and height of the model. Plot one output against the input. State the angle convention.

Week 6: communicate model limits

Explain which findings belong to ideal geometry and which depend on paper thickness, friction or hand accuracy. Suggest one improvement for a rigid-panel version.


Guidance for students and families

Students should draw the crease pattern before and after folding, label points consistently and record mountain–valley conventions. A successful-looking fold is not yet an explanation. Ask which distances, angles or transformations make it work.

Families can encourage low-cost experimentation: compare one precise fold with a hurried one, measure repeating-error growth, or challenge a student to predict the crease before making it. Photograph intermediate stages so the reasoning can be reconstructed.

Avoid turning difficulty into a verdict about “not being mathematical”. Folding combines spatial visualisation, fine motor control and abstract reasoning, and these skills develop at different rates. Using diagrams, transparent paper or a digital geometry tool can make the structure more accessible.


Careers and pathways

Origami mathematics connects to mathematics, mechanical engineering, robotics, architecture, industrial design, materials science, computer graphics and manufacturing. Some pathways emphasise proofs and algorithms; others emphasise prototyping, mechanics or human use.

Useful foundations include geometry, trigonometry, vectors, matrices, calculus, graph theory, programming, physics and experimental design. Artistic judgement and communication also matter because a structure must be understandable and usable.

Learning these ideas can expand options, but a paper project does not guarantee entry to a course or career. It provides a rich way to develop transferable spatial reasoning, modelling and verification.


Did You Know? A fold can draw a parabola without drawing the curve

Repeatedly folding one point to different locations on a line creates creases tangent to a parabolic envelope. Straight lines reveal a curved locus.


Did You Know? Four creases usually need a three-to-one split

At a flat-foldable degree-four interior vertex, Maekawa’s theorem requires the mountain and valley counts to differ by two. That gives three of one type and one of the other.


Did You Know? A backward-looking crease pattern can be an optimisation problem

Designers can begin with a desired deployed surface or motion, then search for crease parameters that minimise mismatch while satisfying physical constraints.


Frequently asked questions

Why is origami mathematical?

Folds implement reflections and geometric constraints. Crease patterns can be analysed with angles, graphs, transformations, kinematics, optimisation and mechanics.

What happens when one point is folded onto another?

The crease is the perpendicular bisector of the segment joining the points because every crease point is equidistant from the point and its reflected image.

What is Kawasaki’s theorem?

For a single ideal flat-foldable interior vertex, alternating sector angles sum to 180°. It is a local condition and does not settle every global issue.

What is Maekawa’s theorem?

At an ideal flat-foldable interior vertex, the number of mountain creases and valley creases differs by two.

What is the difference between a mountain and a valley fold?

They describe opposite crease orientations under a chosen viewing convention. The convention should be stated because flipping the sheet reverses the labels.

Can every crease pattern fold?

No. Sector angles, mountain–valley assignments, layer ordering, rigid-motion compatibility, boundaries and collisions can prevent the intended fold.

Why do engineers care about paper thickness?

Multiple layers and finite bend radii shift hinges, cause interference and change compactness. A zero-thickness model may need offsets or gaps for real materials.

Does origami solve cubic equations?

Certain precisely defined multi-condition folding operations correspond to cubic constructions. This is a mathematical capability of the operation set, not a claim that any informal fold solves any equation.

Is a paper prototype enough for a real product?

No. It can test geometry and motion, but materials, loads, fatigue, safety, tolerances and manufacturing require separate validation.


Useful next reading


Final perspective

Curvature can emerge from many straight creases

A smooth curved surface has continuously changing orientation, while a traditional crease pattern uses flat polygonal faces joined along straight hinges. With enough small faces, the piecewise-flat surface can approximate a curve.

The approximation raises a resolution question. Smaller facets may reduce geometric deviation from the target, but they increase crease count, manufacturing effort and sensitivity to thickness. An objective might measure squared distance between sampled points on the folded surface and a target surface, while a penalty discourages excessive creases.

This is similar to polygonal approximation in computer graphics. More elements can represent finer detail, yet computation and fabrication grow. The best resolution depends on the permitted error and resources.

Gaussian curvature distinguishes bending from stretching

A flat sheet has zero Gaussian curvature away from creases under the ideal model. It can bend into cylinders and cones without stretching, but forming a smooth spherical patch generally requires stretching, compression, cuts or distributed deformation.

Origami concentrates curvature at vertices and creases while keeping faces approximately developable. At a polyhedral vertex, an angle deficit can be calculated as:

K = 2π − Σθᵢ,

where the face angles meeting at the vertex are measured on the folded surface. A positive deficit corresponds to concentrated positive curvature in the polyhedral description.

This gives students a bridge from familiar angle sums to differential geometry. It also explains why wrapping a flat map around a globe without distortion is impossible: intrinsic geometry constrains what bending alone can achieve.

Kirigami introduces cuts as well as folds. A cut changes connectivity and can release geometric constraints, allowing expansion or curvature that an uncut sheet cannot achieve easily.

The mathematics must track new boundaries and possible stress concentration near cut tips. A result for a continuous sheet should not be transferred automatically to a cut lattice. Naming the operation set—folds only, folds plus cuts, rigid panels or elastic sheet—keeps claims precise.

Comparing two designs with the same outline but different cuts is a useful experiment. Measure expansion ratio, recovered shape and damaged regions, while avoiding claims about strength without appropriate testing.

One more connection completes the picture: a fold pattern is both a geometric object and an information system. It must communicate instructions, preserve constraints and tolerate imperfect execution.

Crease-pattern notation is a mathematical language

A diagram may distinguish mountain and valley creases with line styles or colours, mark boundaries, identify hidden folds and number a sequence. The notation works only when its legend is explicit and accessible.

Coordinates supplement the drawing. A crease through (x₁,y₁) and (x₂,y₂) can be reproduced at another scale, while angular and length dimensions support fabrication. Normalised coordinates, expressed as fractions of sheet width, allow one pattern to fit several paper sizes.

Dimensions alone may not determine the folding sequence. Two models can share a final crease pattern but require different orders to avoid trapping layers. Instructions combine static geometry with temporal logic.

Scale changes physical behaviour

If every length is multiplied by s, angles stay the same and areas multiply by s². Geometric similarity is preserved.

Physical behaviour may not scale simply. Sheet thickness might remain fixed rather than multiplying by s. Weight, stiffness and gravity follow different relationships. A tiny paper model and a building-sized panel system can share geometry without sharing mechanics.

Scale models therefore need dimensionless comparisons or carefully justified similarity. Visual resemblance is evidence about form, not automatic evidence about forces.

Sensitivity measures how errors affect the result

Suppose output width W depends on fold angle θ. The derivative dW/dθ describes local sensitivity. A large magnitude means a small angle error can create a large width error.

Students can estimate it with a central difference:

dW/dθ ≈ [W(θ+h) − W(θ−h)]/(2h).

The value depends on units: degrees and radians produce different numerical derivatives. State the angle unit. Designers may prefer a target configuration with lower sensitivity to manufacturing or control error.

Singular configurations can change mobility

A mechanism may reach a configuration where constraints become dependent. At such a singularity, available motions can change. The structure may lock, snap through or gain an infinitesimal motion.

In matrix analysis, a constraint Jacobian can lose rank. A determinant may become zero in a suitable square case, although general systems require broader rank methods.

Singularities are not automatically defects. Designers sometimes exploit them for deployment or stiffness. But motion near one can be sensitive, so validated models and prototypes matter.

Optimisation needs several objectives

A deployable design might seek small folded volume, accurate deployed shape, low actuation force, limited stress and few unique panels. Improving one objective can worsen another.

Multi-objective optimisation can reveal a Pareto frontier. A design lies on it when no objective can improve without worsening at least one other objective under the model.

Decision-makers choose among trade-offs using application requirements. Mathematics organises the options; it does not supply the values automatically.

A reproducible investigation records process

Record paper dimensions and type, crease method, angle convention, measurement device and sequence. Photograph the flat pattern with a scale. Build multiple sheets rather than measuring one object repeatedly as if those were independent constructions.

If another student can reconstruct the design and obtain similar results, the investigation is stronger. Disagreement can reveal hidden assumptions such as crease width, hand pressure or alignment method.

Reproducibility turns a pleasing object into evidence. It is also a habit that transfers directly to mathematics, science, engineering and computing.

Origami shows mathematics changing role as a design develops. Reflection explains a single crease. Angle theorems test a vertex. Graphs record connectivity and layer order. Rotations describe rigid motion. Optimisation searches for a target form. Mechanics and statistics reveal how real materials and tolerances modify the ideal result.

The strongest learner does not merely memorise Kawasaki or fold neatly. They ask what is being modelled, prove the local rule, test the whole pattern, measure disagreement and communicate the limit. A flat sheet then becomes more than paper: it becomes a joyful laboratory for geometry, algorithms, physical reasoning and inventive design.