What Is Logistic Growth?
Exponential growth assumes unlimited resources, which no real system enjoys. Logistic growth fixes this by adding a ceiling. A population grows nearly exponentially while it is small, then slows as it nears the carrying capacity M, the largest size the environment can sustain. The differential equation encodes the idea in one line:
dy/dt = k y (1 − y/M)
When y is tiny the factor (1 − y/M) is close to 1 and growth looks exponential at rate k. As y climbs toward M that factor shrinks toward 0, throttling the growth, and at y = M it stops entirely. Solving the equation by separation of variables and partial fractions gives the closed form
y(t) = M / (1 + A e^(−kt)), with A = (M − y₀) / y₀
whose graph is the famous S-shaped sigmoid.
How to Use the Logistic Growth Calculator
Enter the carrying capacity M, the growth rate k, and the starting value y₀. The calculator assembles the closed-form solution, computes the constant A from your initial condition, and reports the value at any time t you supply. It also finds the inflection point, the moment of fastest growth where the population passes through half of capacity, and plots the S-curve against its ceiling at M and its midline at M/2. Try a starting value above M to watch the curve descend into the ceiling from the other side, which the same formula handles automatically.
Worked Example
A population starts at y₀ = 10 with capacity M = 1000 and rate k = 0.5. The constant is A = (1000 − 10)/10 = 99, so
y(t) = 1000 / (1 + 99 e^(−0.5t))
At t = 10 this gives about 599, already past the halfway mark. The fastest growth happens when y = 500, at time t = ln(99)/0.5, roughly 9.2, where the population is adding members at its peak rate of kM/4 = 125 per unit time. Before that instant growth accelerates; after it, growth decelerates even though the population keeps rising toward 1000. That turning point, visible as the inflection of the S-curve, is the single most important feature of the whole model.
Reading the S-Curve
The logistic curve tells its story in three acts. The first is a lag-and-launch phase where numbers are small and growth is nearly exponential, deceptively slow at first then visibly accelerating. The second is the steep middle, centered on the inflection at y = M/2, where the system adds population fastest. The third is saturation, an ever-slowing approach to the carrying capacity that never quite arrives, since M is a horizontal asymptote. Two equilibria bracket the behavior: y = M is stable, drawing every positive starting value toward it, while y = 0 is unstable, since the smallest seed population departs from it. This structure is exactly what a slope field of the logistic equation reveals at a glance, with all solution curves funneling onto the line y = M.
Common Mistakes to Avoid
- Treating early growth as truly exponential forever. The logistic curve mimics exponential growth only while y is far below M. Extrapolating that early slope leads to wild overestimates, the classic error in naive forecasts of populations and viral spread.
- Placing the inflection at time zero. The fastest growth occurs when the population reaches M/2, not when the clock starts. If y₀ already exceeds M/2, the peak growth moment lies in the past and the curve only decelerates from the outset.
- Assuming the curve reaches M. Capacity is approached asymptotically and never exactly attained in the model. Statements like "the population hits carrying capacity at t = 20" mistake a close approach for arrival.
- Ignoring the case y₀ > M. Starting above capacity is perfectly valid and gives A < 0, producing a curve that falls toward M from above. Overshoot and recovery are real ecological scenarios the formula covers without modification.
Real-World Applications
Logistic growth is one of the most widely reused models in applied science. Ecologists use it for populations limited by food and space, and fisheries management sets sustainable harvest levels from the peak-growth point at M/2, where the stock replenishes fastest. Epidemiology leans on it heavily: in a simple outbreak the fraction infected follows an S-curve, and the inflection is the day new cases peak, the quantity every dashboard tracked during a pandemic. The epidemic chip models exactly this, with capacity 1 representing the whole population.
Beyond biology, the sigmoid describes technology adoption, where each product moves from early adopters through a steep mainstream phase to a saturated market, and businesses time investment against the inflection. Chemical autocatalytic reactions, the spread of rumors and information, language change, and the charging of certain systems all follow logistic dynamics. In machine learning the same S-shaped function, the logistic sigmoid, squashes any real number into a probability between 0 and 1 and sits at the heart of logistic regression and neural network activations, a direct descendant of this differential equation.
Frequently Asked Questions
What is the carrying capacity, physically?
The maximum sustainable size of the quantity, set by limiting resources such as food, space, susceptible hosts, or market size. It is the stable equilibrium the system settles toward and the horizontal asymptote of the S-curve. Above it the population declines, below it the population grows.
Why does the fastest growth happen at half capacity?
The growth rate k y (1 − y/M) is a downward parabola in y, maximized at y = M/2. There the product of "many individuals" and "plenty of room left" is largest. That balance point is the curve's inflection and the peak of the growth-rate bell.
How does logistic growth differ from exponential growth?
Exponential growth has no ceiling and accelerates forever, while logistic growth adds the braking factor (1 − y/M) that halts growth at capacity. For small populations the two agree closely; as capacity nears they diverge sharply, with the logistic curve bending over into its S-shape.
What do the two equilibria mean?
y = 0 and y = M are the constant solutions. The capacity M is stable, attracting nearby solutions, so populations converge to it. The value 0 is unstable, meaning any nonzero start moves away from extinction and toward M. Stability is read from the sign of the growth rate on either side.
Can the growth rate k be negative?
A negative k reverses the roles of the equilibria, making 0 the attractor and driving the quantity toward extinction. Standard logistic growth assumes k > 0, but the formula remains valid and describes decline toward zero when the intrinsic rate is negative.
How do I estimate k and M from data?
The carrying capacity M is the level the data plateaus at, and k is found from the steepness of the middle phase, often by fitting the linearized form ln[y/(M − y)] = kt + constant, which is a straight line in t. Modern practice fits all parameters at once by nonlinear regression, but the plateau and slope give quick estimates.