How MRT Braking Works Using Mathematics

Most passengers notice an MRT train braking only when they feel their bodies lean slightly forward.

But mathematically, the decision to stop began much earlier.

Before the platform is beside us, the train already has a problem to solve: it possesses kinetic energy, it is moving through a railway with a finite distance remaining, there are limits on how strongly it should decelerate, the track may rise or fall, passengers are standing, another train may be ahead, and the train still has to arrive at the correct place at approximately the correct time.

That makes MRT braking far more interesting than “press the brake and stop”.

An MRT does not merely stop at a station. It solves a moving boundary problem in which speed must become zero at the right place, under safety, comfort, energy and timing constraints.

This article is the braking companion to our main pillar, How MRT Timing Works Using Mathematics. Here we zoom in on one part of the journey: the seconds in which motion is deliberately removed.

The RFE — Why Does MRT Braking Mathematics Exist?

The obvious answer is safety.

But that answer is incomplete.

A good braking system must do several jobs at once:

stop the train before it enters unsafe space,
stop at the platform accurately enough for normal passenger operations,
avoid uncomfortable changes in acceleration,
fit the stop inside the timetable,
preserve safe separation from other trains,
recover useful electrical energy where the system allows it,
and remain robust when reality differs from the mathematical prediction.

So the purpose is not simply:

Make velocity = 0

It is closer to:

Reach v = 0
at the correct x
at an acceptable t
with safe separation
within physical limits
without wasting unnecessary energy
while carrying real people

That is the Reason for Existence of the mathematics.

Prompt 1 — Why Must an MRT Start Braking Before the Station?

A moving train cannot lose its velocity instantaneously.

If it is travelling at speed v and can decelerate at magnitude b, the familiar school-level kinematics relationship is:

v² = u² + 2as

For braking to rest, final velocity is zero. If initial speed is v and deceleration magnitude is b, the simplified stopping distance becomes:

d = v² / (2b)

This equation contains one of the most important ideas in railway timing.

Stopping distance grows with the square of speed.

That means doubling speed does not double the idealised braking distance.

It multiplies it by four.

Suppose an illustrative train is travelling at 40 km/h:

40 km/h = 11.11 m/s

Using a deliberately simple constant deceleration of 1.0 m/s²:

d = 11.11² / 2
  ≈ 61.7 m

Now double speed to 80 km/h:

80 km/h = 22.22 m/s

d = 22.22² / 2
  ≈ 246.9 m

The speed doubled.

The simplified stopping distance became four times as large:

61.7 m → 246.9 m

This square relationship explains why speed matters so much to safe separation. A small increase in velocity can consume a surprisingly large amount of stopping space.

It also explains why the station begins to control the train before the station is physically beside us.

If the train needs roughly 200 metres to complete a comfortable braking sequence, the platform is already mathematically relevant while it is still hundreds of metres away.

The future stop creates a constraint on the present speed.

Prompt 2 — Why Is the Simple Braking Formula Not Enough?

The equation:

d = v² / 2b

is useful for understanding the core relationship, but a real train is not a school exercise.

Real braking depends on more than velocity and one constant value for deceleration.

A more realistic conceptual force balance is:

m dv/dt = −Fb − R(v) − mg sin θ

with signs interpreted according to the direction of motion and the track gradient.

Here:

m represents train mass,
Fb is braking force,
R(v) represents rolling, mechanical and aerodynamic resistance,
g is gravitational acceleration,
θ represents the track gradient.

Mass

A peak-hour train carrying many passengers has a larger total mass than the same train carrying fewer passengers.

Newton gives us:

F = ma

For a given required deceleration, a larger mass implies a larger force requirement.

This does not mean a properly designed loaded train suddenly cannot stop. It means mass belongs inside the engineering model.

Gradient

A train braking uphill receives assistance from gravity.

A train braking downhill has gravity acting in the direction of travel.

The gravitational component along a slope is approximately:

Fg = mg sin θ

For small railway gradients, sin θ is small, but the train mass is large. Small geometric slopes can therefore matter.

The same train, at the same speed, does not necessarily have the same braking problem on every piece of track.

Resistance

Train resistance is often approximated using a relationship of the form:

R(v) = A + Bv + Cv²

Resistance therefore changes with speed rather than remaining perfectly constant.

Modern railway models can also include response time, jerk limits, position uncertainty, variations in adhesion, track geometry and safety margins. Recent railway-signalling research expresses approach distance as braking distance plus an explicit safety margin, rather than treating the idealised kinematics answer as the entire safe distance.

This is why the formula from school is important, but only as the first layer.

Prompt 3 — What Is a Braking Curve?

Imagine drawing a graph.

Horizontal axis:

distance remaining to the stopping point

Vertical axis:

permitted speed

Far from the stopping point, the permitted speed may be relatively high.

As remaining distance decreases, the allowable speed must also decrease if the train is still to retain sufficient stopping ability.

A simplified relationship can be written:

vallow ≤ √(2 b deffective)

If we include an abstract safety allowance s:

deffective = dremaining − s

Then the train cannot spend every metre as though it were available for idealised braking.

Some distance is effectively protected against uncertainty.

The result is a braking curve: a mathematical envelope telling the control system how speed must fall as the train approaches a target point.

The target might be a station stopping point, a restrictive movement authority, a section of railway requiring a lower speed, or another operational constraint.

The crucial insight is that signalling does not merely say:

GO
or
STOP

Modern train control can instead create a much more continuous relationship between:

where the train is
+
what is ahead
+
how fast it is travelling
+
how strongly it can safely brake

and from these decide the acceptable speed trajectory.

On Singapore’s North-South and East-West Lines, SMRT states that acceleration, braking and coasting are controlled by the Automatic Train Operation system under CBTC, with algorithms following optimised running curves.

That gives us a real Singapore example of the idea:

The train is not simply being told a destination. Its motion is being shaped as a curve.

Prompt 4 — Why Does Passenger Comfort Change the Braking Mathematics?

If stopping distance were the only objective, we might simply demand the strongest possible braking every time.

But that would make a poor passenger railway.

An MRT contains standing passengers.

When velocity changes, passengers experience acceleration.

a = dv/dt

But comfort is also affected by how quickly acceleration itself changes.

That quantity is called jerk:

j = da/dt

Imagine a train decelerating smoothly at −0.8 m/s².

Now compare that with a train that changes abruptly from zero acceleration to −0.8 m/s².

The final deceleration value may be the same.

The passenger experience is not.

So an idealised comfortable stop can contain phases:

coasting
→ braking force ramps in
→ sustained deceleration
→ braking force eases
→ precise low-speed approach
→ stop

This means the velocity graph, acceleration graph and jerk graph all matter.

For students, this is an excellent example of how derivatives acquire physical meaning:

x(t)  position
↓ derivative
v(t)  velocity
↓ derivative
a(t)  acceleration
↓ derivative
j(t)  jerk

The mathematics does not end at acceleration because the human body notices changes in acceleration too.

RFE: Passenger comfort is not an optional decoration around the mathematics. It changes the allowable trajectory.

Prompt 5 — What Is the Difference Between Normal Braking and a Safety Stop?

A passenger railway has different operating objectives under normal and abnormal conditions.

During ordinary station operation, the preferred stop is controlled, predictable and comfortable.

In a safety-critical situation, the priority hierarchy changes.

The conceptual trade-off might be represented as:

Normal operation:
Safety + punctuality + comfort + energy

Safety intervention:
Safety dominates the objective

This is an important lesson in optimisation.

The objective function is not always fixed.

When the operating state changes, the weights on the objectives can change too.

A conceptual cost function for ordinary driving might be:

J = w1(delay) + w2(energy) + w3(discomfort) + w4(risk)

During a safety intervention, the risk term effectively dominates.

This does not tell us the proprietary logic of any Singapore railway safety system, nor should a public educational article attempt to reproduce such internal rules. The important mathematical idea is enough:

A system can optimise differently when the state of the world changes.

That principle appears far beyond trains: aviation, medicine, power grids, computing, finance and emergency management all use state-dependent operating rules.

Prompt 6 — How Can Braking Save Energy Instead of Only Destroying Motion?

A moving train contains kinetic energy.

The classical expression is:

Ek = ½mv²

Again we see the square of velocity.

Double the speed and kinetic energy becomes four times as large for the same mass.

When a conventional friction brake slows a vehicle, much of that kinetic energy eventually becomes heat.

Electric rail systems can do something more interesting.

During regenerative braking, the traction motors can operate as generators. The train’s kinetic energy is converted back into electrical energy that can be returned to the electrical system and used elsewhere where conditions allow.

LTA states that regenerative braking on Singapore trains can recuperate about 30 per cent of energy consumed and also reduces wear on mechanical braking components. LTA also notes examples where energy from a braking train can be reused elsewhere in the rail system.

So braking is not merely:

motion → waste heat

It can become:

train kinetic energy
→ electrical generation
→ power system
→ useful consumption elsewhere

This creates a second optimisation problem.

Suppose Train A is braking while Train B is accelerating.

In an idealised model, timing these events so that electrical demand and regenerated supply overlap can improve overall energy utilisation.

Academic metro research explicitly studies the synchronisation of train speed profiles and timetables for greater use of regenerative energy.

Now one train’s braking time can matter to another train’s acceleration.

The braking problem has escaped the train and entered the network.

Prompt 7 — How Does Braking Connect to Signalling and Headway?

A train’s braking ability affects how closely trains may safely operate.

This connects stopping distance to headway.

Let Train A be ahead and Train B behind.

Train B needs sufficient protected space and an appropriate permitted-speed profile so that it can respect the safe movement authority ahead.

At a simplified conceptual level:

safe separation
≥ braking requirement
+ uncertainty allowance
+ system protection margin

If train position is known more precisely and the signalling system can continuously manage movement authority and speed, trains may be able to operate with less unused space than a coarser control arrangement while still maintaining the required safety envelope.

This is part of the reason Communications-Based Train Control matters for high-frequency metro operation.

On the North-South and East-West Lines, SMRT describes the CBTC system as controlling acceleration, braking and coasting through Automatic Train Operation. The same programme now analyses movement and energy-use profiles to recommend improvements in timetables and operations.

Braking therefore connects to:

speed
→ stopping distance
→ safe separation
→ headway
→ trains per hour
→ passenger capacity

This is why a braking equation belongs inside a city-capacity equation.

A simple headway illustration

If a railway could operate trains at a 120-second headway, the theoretical flow is:

3600 / 120 = 30 trains per hour

At 100 seconds:

3600 / 100 = 36 trains per hour

The 20-second reduction represents a theoretical 20 per cent increase in train paths per hour.

Of course, real line capacity is also constrained by stations, terminals, junctions, passenger loading, fleet availability, power and recovery margin. But the braking and signalling problem is one of the foundations beneath high-frequency operation.

Prompt 8 — Can We Reconstruct One Braking Event Mathematically?

Yes, as long as we are clear that this is an educational example and not an actual Singapore MRT braking curve.

Suppose a hypothetical train is travelling at:

72 km/h

Convert to metres per second:

72 / 3.6 = 20 m/s

Suppose our simplified comfortable braking phase uses a constant magnitude:

b = 0.8 m/s²

The idealised braking time is:

t = v / b
  = 20 / 0.8
  = 25 s

The idealised braking distance is:

d = v² / 2b
  = 20² / 1.6
  = 250 m

So under this deliberately simplified model, braking begins around a quarter of a kilometre before the target.

Now change only one thing.

Increase initial speed from 72 km/h to 81 km/h.

81 km/h = 22.5 m/s

Then:

d = 22.5² / 1.6
  ≈ 316.4 m

Speed increased by:

12.5%

but idealised braking distance increased from 250 metres to about 316 metres:

increase ≈ 26.6%

This is the square law at work.

Now introduce a downward gradient, response allowance, comfort-limited jerk and a safety margin.

The true engineering curve moves again.

Now introduce a speed restriction before the station.

The curve changes again.

Now place another train ahead.

The permissible movement changes again.

Now demand the same arrival time while reducing energy consumption.

The optimisation changes again.

This is why a real braking event is better understood as a continuously constrained trajectory than as one number called “braking distance”.

The Three Different Distances Students Should Not Confuse

When thinking mathematically about braking, it helps to distinguish three concepts.

1. Idealised physics stopping distance

This is the clean textbook answer under simplified assumptions:

dphysics = v² / 2b

2. Engineering braking distance

This can account for real train characteristics, gradients, resistance, brake response and other engineering variables.

3. Safe signalling distance or movement envelope

This must also account for uncertainty and required protection margins.

Therefore:

textbook stopping distance
≠ complete engineering braking model
≠ complete safe signalling envelope

This distinction matters enormously when students move from mathematics exercises into engineering.

Why a Train Sometimes Seems to Brake Early

A passenger may occasionally feel that the train is slowing “too early”.

But an early reduction in traction or speed can have several mathematically sensible explanations:

the train may be following an energy-efficient running curve,
it may be coasting before the braking phase,
the permitted speed may be lower further ahead,
the movement authority may require a lower speed profile,
the timetable may not require the train to reach the next station any earlier,
the system may be preserving regular headway rather than maximising one train’s speed.

SMRT’s Green CBTC programme is a useful public example: the Automatic Train Operation system can use optimised running curves and predetermined coasting locations to save traction energy with negligible impact on travel time.

So what feels like “doing less” can be the optimal decision.

The Energy Mathematics Hidden in One Stop

Consider a simplified train of mass m travelling at speed v.

Kinetic energy is:

Ek = ½mv²

Suppose the train mass is represented only for illustration as 200,000 kg and speed is 20 m/s.

Ek = ½(200,000)(20²)
   = 40,000,000 J
   = 40 MJ

That is a large amount of mechanical energy in one deliberately simplified example.

If some portion can be converted back into useful electrical energy, repeated braking across thousands of station stops becomes an important system-level efficiency opportunity.

This is why LTA’s statement that regenerative braking can recuperate about 30 per cent of energy consumed matters so much. The value is not one spectacular stop. The value is repetition.

small saving per event
× many trains
× many stations
× many operating hours
× many days
= large network effect

The Mathematics of Repetition

This is one of the most useful infrastructure lessons.

Suppose an improvement saves only E_s units of energy per stop.

If there are n relevant braking events per day:

Edaily = n Es

Over 365 days:

Eyear = 365 n Es

Network infrastructure often becomes powerful through multiplication.

SMRT reported that Phase 1 of its Green CBTC work on the North-South and East-West Lines achieved an additional 8 per cent energy saving for automatic trains, amounting at the time to more than 15 million kWh annually. SMRT’s current decarbonisation information places Phase 1 annual savings at around 17 GWh.

That is what happens when tiny timing and trajectory choices are repeated across a large railway.

The Braking Paradox

Braking seems like the opposite of progress.

The train is losing speed.

Yet good braking is what allows high-speed movement beforehand.

If you cannot guarantee that motion can be safely removed, you cannot safely permit the motion in the first place.

So:

better stopping knowledge
→ better speed authority
→ tighter control
→ more reliable headway
→ more useful capacity

The ability to stop is part of the ability to go.

A railway is fast only because it knows how to become slow.

What Students Can Learn from MRT Braking

MRT braking connects many mathematical ideas that are often taught separately.

Algebra: rearranging braking equations.
Quadratic relationships: stopping distance and kinetic energy both contain v².
Kinematics: relating speed, time, distance and acceleration.
Calculus: velocity and acceleration as derivatives.
Differential equations: modelling force and motion continuously.
Optimisation: balancing time, energy, comfort and safety.
Probability: representing uncertainty rather than pretending every parameter is exact.
Control theory: comparing desired and observed motion and correcting continuously.
Systems mathematics: connecting one train’s braking to headway, capacity and network energy.

The railway does not ask which chapter of the textbook the equation came from.

It simply needs the equations to work together.

RFE Return — What Is a Good Stop?

A good MRT stop is almost invisible.

Passengers rarely applaud it.

The train simply arrives.

But that ordinary arrival means many things went right.

The train knew enough about where it was.
The permitted-speed curve remained safe.
Braking began early enough.
Deceleration remained within the required operating envelope.
The stop was comfortable enough for passengers.
The train reached the right place.
The timing remained useful to the timetable.
Where possible, part of the train’s kinetic energy was recovered rather than wasted.

The passenger sees doors opening.

The mathematics sees a boundary condition satisfied:

v(tf) = 0
x(tf) = xtarget

with all the constraints respected along the path.

That is a much more beautiful definition of stopping.

Conclusion — The Station Reaches Backwards Through Time

When an MRT enters a station, the platform appears to be in front of the train.

Mathematically, however, the platform has been influencing the journey for hundreds of metres.

The required final state:

position = platform stop
velocity = 0

reaches backwards through the calculation and limits what the train may do now.

That is why braking is not merely an action at the end of a journey.

It is part of the entire journey.

Speed determines kinetic energy.

Speed determines stopping distance.

Stopping distance influences permitted speed.

Permitted speed influences headway.

Headway influences capacity.

Braking energy influences electricity consumption.

Passenger comfort influences the acceptable acceleration curve.

And uncertainty influences how much margin the system must preserve.

Speed
→ energy
→ braking
→ safe distance
→ signalling
→ headway
→ capacity
→ passenger movement
→ city

One ordinary stop connects all the way back to the operation of Singapore.

The train stops because the mathematics began before we noticed it.

Key Equations

v² = u² + 2as            Kinematics

d ≈ v²/(2b)              Simplified stopping distance

t ≈ v/b                   Simplified braking time

Ek = ½mv²                 Kinetic energy

F = ma                    Force and acceleration

Fg = mg sin θ             Gradient component

R(v) = A + Bv + Cv²       Generic train-resistance model

v = dx/dt                 Velocity

a = dv/dt                 Acceleration

j = da/dt                 Jerk

vallow ≤ √(2bdeffective)  Simplified braking-speed envelope

Important: These equations are used here for education and conceptual modelling. Actual railway braking and signalling systems use validated train-specific models, operating rules, safety margins, system measurements and engineering verification that are much more detailed than these classroom examples.

Continue the Series

Read the main pillar: How MRT Timing Works Using Mathematics: Inside the Seconds That Keep Singapore Moving.

The next mathematical layer is headway: why 100 seconds and 120 seconds can imply very different passenger capacity even when the trains themselves do not become faster.

How MRT Braking Works Using Mathematics: Why a Train Starts Stopping Before the Station