Static and dynamic optimization differ in how time is represented in a decision problem. A static optimization model determines a decision for a specified situation without explicitly modelling how decisions evolve over time. It may still use data measured over a period, but it does not represent a sequence of linked states and decisions. Dynamic optimization considers decisions over time, often with the state of the system changing from one period to another. Earlier decisions may affect future resources, costs, or opportunities. Dynamic programming is one important technique for solving certain dynamic optimization problems. Other approaches include optimal control and multi-period mathematical programming. Dynamic models are useful when timing, future consequences, and the relationship between present and future decisions are essential.
Static and Dynamic Optimization
Static optimization determines the best decision for a single modelled situation, while dynamic optimization considers decisions and system changes across multiple time periods.
Static: minimize f(x)
Dynamic system: xₜ₊₁ = F(xₜ, uₜ)
Multi-period objective: minimize Σₜ₌₀ᴺ⁻¹ cₜ(xₜ, uₜ) + cᴺ(xᴺ)
Explanation
Variables
x: static decision vector
xₜ: system state at time t
uₜ: decision or control at time t
F: state-transition function
cₜ: cost at time t
N: number of time periods
Example or application
A company choosing how many products to manufacture in one fixed period may use a static model. A company planning production, inventory, and purchasing over the next 12 months may need a dynamic model.
Definition
Static optimization finds an optimal decision for a single modelled state or period, whereas dynamic optimization determines decisions in a system whose state or choices evolve over time.
Purpose
The purpose is to choose a modelling framework that reflects whether the problem is a single decision or a sequence of connected decisions over time.
Importance
Dynamic optimization is essential when present decisions influence future system states, resource availability, or performance.
Learning Objectives
Define static and dynamic optimization.
Explain how time affects decision-making.
Interpret state variables and control variables.
Understand state-transition relationships.
Identify suitable applications of dynamic optimization.
Key Concepts
Static optimization
Dynamic optimization
Time period
System state
Control variable
State transition
Dynamic programming
Working Principle
Static optimization evaluates a decision within one specified model. Dynamic optimization links decisions across time through state-transition equations and an objective that accounts for current and future consequences.
Process
Determine whether the problem involves one decision period or multiple linked periods.
Define the relevant state variables.
Identify the decisions available at each period.
Describe how decisions affect future states.
Define the total objective across the required time horizon.
Apply an appropriate dynamic solution method.
Evaluate the resulting sequence of decisions.
Applications
Multi-period production planning
Inventory replenishment
Energy storage and dispatch
Investment planning
Optimal control systems
Water-resource management
Worked Example
A simple storage system begins with 10 units of inventory. At each period, new stock can be purchased and demand reduces inventory. The state equation is xₜ₊₁ = xₜ + uₜ − dₜ, where xₜ is current inventory, uₜ is the quantity purchased, and dₜ is demand. A dynamic optimization model can minimize total purchasing and storage costs over several periods while ensuring inventory never becomes negative.
Common Misconceptions
Static optimization does not necessarily mean the problem is solved instantly; it means time-dependent state evolution is not explicitly represented.
Dynamic optimization is not simply repeating the same static optimization independently for every period.
Current decisions may affect future feasible choices.
Dynamic programming is one method for dynamic optimization, not the only method.
Practice Questions
Define static and dynamic optimization.
What is a state variable?
How can an inventory decision in one period affect the next period?
Explain the role of a state-transition equation.
Give two examples of dynamic optimization.
Key Takeaway
Static optimization focuses on a specified decision situation, while dynamic optimization accounts for linked decisions and changing system states over time.
Prerequisites
Functions and sequences
Mathematical Modelling in Optimization
Basic algebra
Reference
Bellman, R. Dynamic Programming.
Bertsekas, D. P. Dynamic Programming and Optimal Control.
Related Topics
Deterministic and Stochastic Optimization
Continuous and Discrete Optimization
Mathematical Modelling in Optimization
Optimization Problem Formulation
Sources and verification
Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.
- Source verification pending; check an authoritative textbook, university, standards body, or peer-reviewed source.
Suggest a correction or source · More in Optimization & Operations Research