Blog main article:
Municipal AI—From Local Optima to the Global Optimum
Why the Best Individual Decision Does Not Automatically Lead to the Best Overall Outcome
Local government decisions are rarely made under ideal conditions. They are made under time pressure, with limited budgetary resources, conflicting political goals, and increasing complexity.
Cities and municipalities must simultaneously invest in schools, infrastructure, digitalization, climate protection, social services, and public safety. However, available budget funds are often insufficient to implement all projects.
In practice, this leads to a familiar pattern: Individual projects are evaluated on their merits, prioritized politically, and then included in the budget.
But even if every individual decision seems reasonable, the resulting overall portfolio is by no means necessarily optimal.
The best project at the local level is not automatically part of the best project portfolio overall.
This is precisely where the strategic importance of mathematical portfolio optimization with StratePlan comes into play.
1. Local Optima: Plausible, but Systematically Incomplete
In traditional municipal budgeting and investment processes, projects are often considered individually at first:
- Project A makes economic sense.
- Project B is urgently needed from a technical standpoint.
- Project C aligns with a political priority.
- Project D promises a significant social impact.
Each of these assessments may be correct on its own.
The real challenge, however, arises from the interplay between the projects.
- Projects compete for the same budget funds.
- Funding programs alter actual funding needs.
- Project dependencies affect feasibility.
- Time constraints limit implementation.
- Different combinations produce different overall effects.
Viewing projects in isolation can therefore lead to a selection that, while seemingly plausible, does not make the best possible use of the available budget.
A locally good result does not necessarily constitute a globally optimized portfolio.
2. Why Municipal Investment Decisions Become Exponentially Complex
As soon as multiple projects are under consideration simultaneously, the number of possible project combinations grows exponentially.
In a simplified model, each project can either be implemented or not implemented.
Thus, with N independently selectable projects, there aretheoretically 2N possible combinations.
- 10 projects: 1,024 combinations
- 20 projects: 1,048,576 combinations
- 30 projects: 1,073,741,824 combinations
- 40 projects: 1,099,511,627,776 combinations
- 50 projects: 1,125,899,906,842,624 combinations
With just 50 projects, this results in a theoretical decision space of approximately 1.126 quadrillion possible project combinations.
Not every combination is permissible due to budget constraints, legal requirements, or project dependencies. Nevertheless, the sheer scale illustrates the mathematical challenge.
A manual review or the comparison of a few budget scenarios cannot systematically map this decision space.
Even traditional spreadsheets and simple prioritization methods reach practical limits when dealing with complex combinatorial problems. Specialized optimization methods, on the other hand, can specifically search for particularly high-performing project combinations.
The crucial question, therefore, is not which individual project promises the greatest benefit, but which combination of all available projects achieves the best overall result under the given constraints.
3. The Key Fallacy: Prioritizing Projects Instead of Optimizing Portfolios
Prioritization and optimization are two distinct tasks.
In prioritization, projects are evaluated based on specific criteria and ranked.
Portfolio optimization, on the other hand, examines which combination of projects achieves the highest defined overall value, taking into account available resources and other constraints.
A simple example illustrates the difference.
A city has an investment budget of 10 million euros.
- Project A: Cost €7 million, benefit value 90
- Project B: Cost €5 million, benefit value 75
- Project C: Cost €5 million, benefit value 70
If projects were prioritized solely based on the highest individual value, Project A would come first.
However, the remaining budget of 3 million euros is not sufficient for either of the other two projects.
The total benefit value is thus 90.
If, on the other hand, Projects B and C are implemented together, the budget is fully utilized and a total benefit value of 145 is achieved.
In this simplified example, the portfolio consisting of B and C achieves a total benefit value that is approximately 61% higher than selecting Project A alone.
The figures are for illustrative purposes only. However, they demonstrate a fundamental principle:
The best combination does not necessarily have to include the project with the highest individual score.
4. From Subject-Matter Prioritization to Mathematical Calculation
With StratePlan, municipal investment projects can be viewed as a cohesive portfolio and mathematically optimized.
For a simple municipal decision-making model, only three pieces of information are initially required per project:
- Project ID: Unique identifier for the project
- Expenses: Planned investment requirements
- Weight: Relative urgency or importance as determined by subject matter experts
The weight can range, for example, from 0.01 to 0.99.
The higher the weight, the greater the relative importance of the project within the agreed-upon evaluation model.
The subject-matter departments provide the decisive evaluation
Weights should not be arbitrarily determined by software.
They are determined in advance by the relevant departments and their specialists.
The Department of Education understands the urgency of school construction projects. The Department of Public Works knows the condition of municipal infrastructure. The fire department assesses its technical and operational requirements. The Department of Social Services evaluates the need for social services facilities.
The department specialists possess the necessary expertise to assess the urgency of their projects.
To ensure these assessments are comparable, the Finance Department should provide a standardized, cross-departmental evaluation grid.
StratePlan then conducts the mathematical analysis of the possible project combinations.
This results in a clear division of responsibilities:
- Line departments: Evaluate projects and determine weightings
- Finance Department: Define budget constraints and framework conditions
- StratePlan: Mathematically optimize project combinations
- Finance Committee and City Council: Compare scenarios and make political decisions
5. Practical Example: 50 Projects, 81 Million Euro Budget
Let’s consider an example German city with the following initial data:
- 50 planned investment projects
- 220 million euros in total investment needs
- 81 million euros in available investment budget
The city can therefore finance only a portion of the planned measures.
The relevant departments have evaluated the projects in advance and assigned weights to them.
The Finance Department enters this data into the optimization model.
StratePlan then calculates a project combination, taking into account the available budget and the specified evaluation parameters.
The result shows which projects are included in the calculated portfolio, what expenses will be incurred, and what overall rating the portfolio achieves.
This transforms a list of competing investment requests into a decision-making space that can be analyzed mathematically.
6. Live budget simulation in the Finance Committee or City Council
Mathematical portfolio optimization becomes particularly interesting when it is directly integrated into the political decision-making process.
Let’s imagine a meeting of the Finance Committee.
The treasury department presents the initial calculation for the investment budget of 81 million euros.
Now a member of the committee asks:
“What happens if we prioritize our schools more highly?”
For example, the weighting of selected school projects is increased from 0.75 to 0.95.
StratePlan recalculates the entire project portfolio.
The finance committee can then see:
- Which additional school projects are included?
- Which other projects might be dropped?
- How does the overall evaluation of the portfolio change?
- How much budget is needed?
- What conflicting goals arise from the changed prioritization?
Another council member asks:
“What would be the impact if we gave greater weight to municipal infrastructure instead?”
The parameters are adjusted again, and another calculation is performed.
Budget changes can also be simulated:
- What changes with a budget of 75 million euros?
- What additional projects would be possible with a budget of 90 million euros?
- What combination results when the weightings remain unchanged?
- Which projects would remain part of the portfolio even if political priorities changed?
Change the weightings. Recalculate. Compare scenarios. Understand the implications.
This allows a finance committee to discuss different political objectives based on their specific impacts on the entire investment portfolio.
7. Sensitivity Analysis: How stable are municipal investment decisions?
A key advantage of mathematical budget simulations is that they allow for more than just a single result to be calculated.
By making targeted changes to weightings and budget constraints, different scenarios can be examined.
These sensitivity analyses answer important questions:
- Which projects are selected in nearly all scenarios?
- Which projects are particularly sensitive to changes in weightings?
- Which priorities lead to the greatest changes in the portfolio?
- Which projects are in direct competition for the same budget funds?
- How do increased or reduced budgets change the project mix?
A project that is selected under numerous different assumptions can be considered particularly robust within the model under consideration.
Other projects may only be considered under specific policy priorities.
This reveals which decisions remain stable in the face of changing assumptions and which are particularly dependent on political priorities.
8. What is not systematically calculated remains an incomplete selection
In traditional budget deliberations, only a few alternatives are often compared with one another.
For example, the original budget proposal, a revised version from the administration, and individual amendment proposals from the political caucuses.
These options may be carefully crafted. Nevertheless, when dealing with a large number of projects, they represent only a very small fraction of the theoretical decision space.
With 50 projects, there are theoretically more than one quadrillion combinations to choose from.
Therefore, considering five or ten budget variants alone does not provide mathematical proof that the best permissible combination of projects has been found.
This is precisely where specialized optimization algorithms come into play.
StratePlan uses mathematical and AI-based optimization methods to efficiently explore large combinatorial decision spaces and calculate high-performing project portfolios.
Depending on the problem and the method used, an optimal or approximately optimal solution can be determined. A mathematically proven global optimum requires a corresponding proof of optimality.
The key advancement lies in not limiting municipal investment decisions exclusively to a few manually developed variants, but rather in systematically examining the decision space mathematically.
9. Global Optimum: Mathematically Defined, Politically Accountable
A global optimum is not a political opinion.
It refers to a valid solution that achieves the best possible objective function value within a fully defined mathematical model.
Such a model requires clearly defined objectives and constraints.
These include, for example:
- The available investment budget
- The technical weightings of the projects
- Legal and contractual obligations
- Dependencies between individual projects
- Available personnel and technical resources
- Temporal and organizational constraints
Under these assumptions, mathematical optimization can determine the best possible permissible combination of projects or approximate it using suitable methods.
It should be noted that several different portfolios can achieve the same optimal objective function value.
Likewise, the result always depends on the previously defined weightings and constraints.
The mathematics within the model is transparent. Defining political goals remains a human and democratic task.
StratePlan therefore replaces neither the city council nor the finance department nor the specialized administrative departments.
It complements their work by providing a mathematical basis for decision-making.
10. From Prioritization to Optimization: A New Role for the Finance Department
The traditional finance department consolidates budget proposals, assesses financial flexibility, and prepares decisions for the political bodies.
With mathematical portfolio optimization, this task can be expanded to include an additional strategic function.
The finance department becomes the coordinator of a dynamic budget and investment simulation.
It consolidates the technical assessments from the various departments, defines the available funds, and prepares alternative budget scenarios.
In the finance committee or city council, these scenarios can then be mathematically analyzed and compared with one another.
This shifts the central focus of municipal budget planning.
Instead of asking exclusively:
“Which projects can we afford?”
we can also ask:
“Which combination of our projects achieves the highest defined overall impact with the available funds?”
And furthermore:
“How does this combination change if we adjust our policy priorities or financial constraints?”
11. Municipal AI Does Not Mean the Automation of Political Decisions
The use of AI in municipal budget planning must not be confused with an automatic transfer of political responsibility to software.
StratePlan serves the purpose of ex-ante optimization: the mathematical analysis of possible decisions before they are made at the political level.
The relevant administrative departments provide their expertise.
The finance department defines the financial and organizational framework.
Mathematical optimization examines possible project combinations.
The democratically elected bodies decide on the actual implementation.
The weightings, constraints, and results used should be documented so that the calculations remain transparent and verifiable.
AI does not assume political responsibility. It expands the mathematical foundation upon which political decisions can be made.
12. Conclusion: From the Best Local Project to the Best Possible Overall Portfolio
Municipal budget planning is not merely a matter of limited financial resources.
Above all, it is a complex selection process involving numerous competing projects.
The greater the number of possible investments, the more difficult it becomes to assess their interactions and possible combinations using manual methods alone.
The decisive shift, therefore, lies in the transition from isolated project prioritization to mathematical portfolio optimization.
The departments evaluate. The finance office coordinates. StratePlan calculates. The city council decides.
Using an interactive budget simulation, political bodies can model different priorities, adjust budget parameters, and compare the resulting project portfolios.
This enables a new level of quality in municipal decision support:
- Systematic analysis of complex project combinations
- More efficient use of limited budget resources
- Transparent comparison of policy scenarios
- Transparent presentation of conflicting objectives
- More informed investment decisions
The future of municipal budget planning lies not only in better prioritizing projects, but also in calculating their optimal combination.
StratePlan makes complex municipal decision-making spaces mathematically analyzable, thereby creating an additional foundation for transparent, impact-oriented budget decisions.
How much more impact can you get from your municipal investment budget?
See for yourself how municipal investment projects can be mathematically optimized and how different budget scenarios can be compared.
Same projects. Same budget. Mathematically optimized combinations.