1. Working with Agentic AI: Operations Research Analyst

How human skills and agentic skills are dynamically integrated and coordinated across operations research responsibilities

1.1. Operations Research Analyst as an Occupation

An operations research analyst applies advanced mathematical and analytical methods to help organizations solve complex problems, allocate resources efficiently, and make better decisions. This occupation connects mathematical abstraction, systems engineering, and business strategy by transforming real-world operational constraints and objectives into rigorous quantitative models. Unlike purely descriptive or predictive analytical roles, operations research is fundamentally prescriptive, focusing on optimization, simulation, and decision analysis to determine the best possible course of action. To ensure these mathematical models accurately reflect operational realities and yield implementable solutions, the analyst must understand how physical constraints, system dynamics, uncertainty, organizational objectives, and computational limitations collectively determine the validity and usefulness of the modeled outcomes.

Typical responsibilities of an operations research analyst include:

• problem formulation and mathematical modeling;
• optimization and prescriptive analytics;
• simulation and stochastic modeling;
• decision analysis and risk evaluation;
• model validation, sensitivity analysis, and testing; and
• operational implementation and decision support.

These responsibilities are fulfilled through distinct tasks and subtasks, each requiring one or more actions with specific skill requirements. Human skills and agentic skills both contribute to these responsibilities, but their levels and forms of involvement differ according to the responsibility, task, action, data, operational context, technical environment, and expected outcome.

1.2. Human Skills and Agentic Skills in Operations Research

Operations research increasingly integrates both human skills and agentic skills.

Human skills are reusable capabilities possessed and developed by operations research analysts. They include mathematical abstraction, systems thinking, operational intuition, constraint definition, algorithmic reasoning, critical evaluation, tradeoff analysis, and stakeholder communication.

Agentic skills are reusable capabilities available to AI agents through their underlying models, contextual information, instructions, constraints, tools, and computational resources. In operations research, these skills include mathematical programming code generation, simulation scripting, automated parameter sweeping, solver configuration, algebraic manipulation, literature retrieval for algorithmic approaches, data preprocessing, and diagnostic summarization.

Both skill types contribute to the same occupational responsibilities, but their involvement varies by action. Translating a loosely defined operational bottleneck into a rigorous mathematical framework relies primarily on human knowledge of system dynamics and physical constraints; an AI agent then uses agentic skills to draft the initial algebraic equations, generate the solver code, and format the data inputs. In optimization, the analyst first determines the appropriate objective functions and hard constraints, and an AI agent subsequently writes the optimization scripts, configures the solver parameters, and extracts the shadow prices. For simulation, an AI agent may generate the discrete-event logic, while the analyst interprets the output distributions to ensure they reflect real-world stochastic behavior. Across these actions, human and agentic skills are activated sequentially or iteratively based on task requirements, with the human professional maintaining oversight and final judgment.

The relationship between these skills is not a fixed division where certain responsibilities permanently belong to humans and others to AI. Instead, human and agentic involvement shifts across tasks and between different actions within the same task.

1.3. Dynamic Human–Agentic Skill Integration and Collaboration in Operations Research

Operations research functions can be understood as an occupation where responsibilities are fulfilled through tasks and subtasks, tasks and subtasks are executed through actions, and actions require appropriate configurations of human and agentic skills.

An occupation provides the overall professional context. For operations research, this context encompasses formulating complex problems mathematically, optimizing system performance, simulating dynamic environments, and supporting high-stakes operational decisions.

A responsibility represents an expected area of professional work or an outcome that the operations research analyst accomplishes or maintains. Responsibilities such as mathematical modeling, optimization, simulation, and decision support each contain multiple tasks.

A task is a context-dependent unit of work used to fulfill part or all of a responsibility. When a task is too broad or complex for reliable execution, it is decomposed into smaller subtasks.

Tasks and subtasks are executed through one or more actions. Actions represent reusable and configurable operations such as defining a decision variable, writing a linear constraint, configuring a mixed-integer programming solver, generating a Monte Carlo distribution, executing a discrete-event simulation, calculating an expected monetary value, or extracting sensitivity ranges.

Each action has skill requirements. These requirements describe the capabilities needed for effective and reliable execution. Depending on the action and its context, human skills, agentic skills, or an integration of both can satisfy these requirements.

The skills available for operations research reside in a skill space containing both human skills and agentic skills. Relevant skills in this space can serve as candidate skills for executing the actions required by a particular task. A human skill in systems thinking, for example, can serve as a candidate for identifying hidden feedback loops in a supply chain network, while an agentic skill in simulation scripting can serve as a candidate for coding those loops into a dynamic model.

Candidate skills are evaluated relative to the requirements of the action and its occupational and task context. Appropriate skills are then selected and organized into a skill configuration for execution. Depending on these requirements, a configuration may rely mainly on human skills, mainly on agentic skills, or on an integrated arrangement of both.

1.3.1. Skills Competition and Skills Collaboration

Two important relationships can occur between human skills and agentic skills when constructing a skill configuration: skills competition and skills collaboration.

Skills competition occurs when both a human skill and an agentic skill can satisfy the same or similar requirements for an action, but simultaneous application of both is unnecessary. Under a particular occupational and task context, one skill type can provide a better fit for the action than the other.

For example, both an operations research analyst and an AI agent can calculate the critical path and baseline resource allocation for a standard project scheduling network. When the task dependencies, durations, and resource limits are clearly specified and mathematically straightforward, the agentic skill provides fast and scalable execution. When the schedule requires modeling complex, non-linear human fatigue factors, ambiguous regulatory constraints, or strategic resource hoarding behaviors, the analyst’s human skills provide the necessary mathematical abstraction and operational understanding.

The purpose of skills competition is not to establish whether human or agentic skills are generally superior. Rather, it identifies which available skill provides the better fit for a particular action under particular conditions.

Skills collaboration occurs when one skill type alone cannot effectively satisfy the requirements of a task or its actions, or when combining complementary human and agentic skills produces stronger task performance.

Collaboration does not require human and agentic skills to operate simultaneously. Instead, they are activated at different stages of action execution in a sequential, iterative, or recursive manner according to evolving task requirements.

For example, an analyst first identifies that a multi-echelon distribution network is experiencing severe localized stockouts and determines which operational constraints require mathematical representation. An AI agent then drafts the initial mixed-integer linear programming (MILP) formulation, generates the solver code, and executes a baseline optimization. The analyst subsequently reviews the solver’s output, identifies that the model is allowing physically impossible teleportation of inventory between nodes due to a missing transit-time constraint, and formulates the corrective mathematical logic. Based on the analyst’s direction, the agent next updates the code, applies a heuristic relaxation to handle the increased computational complexity, and re-runs the solver. Finally, the analyst evaluates the resulting network design and determines which structural changes are operationally feasible to communicate to logistics managers. Throughout this sequence, human and agentic skills are activated iteratively as task requirements evolve, while the analyst directs the investigation and evaluates its outcomes.

This dynamic relationship allows operations research analysts to leverage the complementary strengths of both skill types rather than relying on a permanent allocation of work.

1.3.2. Skill Configuration and Task Execution

A skill configuration specifies which human skills and agentic skills are used to execute the actions associated with a task or subtask and how those skills are coordinated.

The configuration depends on factors such as action requirements, mathematical complexity, computational tractability, data availability, system uncertainty, expected reliability, available solvers, information sensitivity, decision consequences, stakeholder expectations, and intermediate results.

For one action, an agentic skill may independently satisfy the requirements. For another, a human skill may provide the better fit. A complex task may require several configurations across multiple actions, with human and agentic skills repeatedly interacting as the task progresses.

The configuration is therefore dynamic rather than permanent. It may change when the action changes, new operational data becomes available, an agentic result fails to satisfy computational limits, or human interpretation identifies a new structural flaw in the model.

Throughout this process, the operations research analyst remains the professional responsible for directing and controlling the work associated with the occupation and its responsibilities. Agentic AI provides powerful complementary capabilities that expand what the professional can formulate, code, simulate, optimize, test, and communicate.

1.3.3. Task Performance Evaluation and Skill Space Adaptation

The performance produced through a selected skill configuration is evaluated against predefined evaluation metrics. These metrics determine whether the human skills, agentic skills, or their integrated configuration are sufficiently capable of performing the relevant task and its required actions.

Evaluation criteria depend on the work being performed. In operations research, they include mathematical correctness, computational tractability, operational validity, constraint adherence, robustness to uncertainty, reproducibility, interpretability, and relevance to business requirements.

When a selected skill or skill configuration produces performance above the defined threshold, it passes the performance evaluation and is retained for future applications under appropriate conditions. Repeated successful use provides additional evidence about where that skill or configuration performs effectively.

When performance falls below the required threshold, the corresponding skill or configuration is not reused without modification. The relevant human or agentic skills may be refined or reconfigured, existing skills can be combined differently, or new human and agentic skills may be developed or constructed from scratch when the existing skill space lacks adequate capability.

For example, an AI agent first generates optimization code for a complex routing problem, but the resulting model repeatedly fails to converge within acceptable time limits due to poor mathematical formulation. Task performance evaluation identifies this computational intractability as falling below the required threshold. The agentic skill is then refined through instructions to apply specific decomposition techniques, stronger contextual prompts regarding symmetry-breaking constraints, or integration with specialized algorithmic libraries. The revised skill is evaluated again, and once it satisfies the required performance criteria, it is retained for appropriate future routing tasks. Human skills follow a corresponding development process: when an analyst lacks the knowledge required to apply a newly developed stochastic programming technique to a complex energy grid problem, the relevant capability is developed through learning, collaboration, and practical experimentation before its performance is evaluated through subsequent work.

Refined and newly constructed skills that satisfy the required performance criteria are added to the skill space. In this way, task performance evaluation provides feedback not only for completing current work but also for continuously optimizing and expanding the available skill space.

The skill space is not constructed as an isolated collection of occupation-specific representations. Many human and agentic skills used in operations research are reusable across occupations. Mathematical reasoning, code generation, statistical analysis, simulation logic, literature retrieval, summarization, visualization, and documentation contribute to work in data science, industrial engineering, financial engineering, logistics, economics, and many other fields.

Reusable skills can therefore be represented once and configured differently according to the responsibilities, tasks, actions, and contexts of different occupations. This allows the skill space to serve a growing range of occupations without unnecessarily duplicating or exploding the number of skill representations. Occupational specialization is introduced through task context, action requirements, parameters, data sources, tools, constraints, and skill configurations rather than by creating a completely separate skill representation whenever the same underlying capability appears in another occupation.

The resulting process is adaptive: occupational responsibility → task and subtask → actions → skill requirements → candidate human and agentic skills → skill configuration → task execution → task performance evaluation → skill refinement, reconfiguration, or construction → optimized and expanded skill space

This cycle allows human and agentic capabilities to evolve alongside the requirements of operations research work.

1.4. Essential Responsibilities of Operations Research Analyst

1.4.1. Problem Formulation and Mathematical Modeling

Problem formulation and mathematical modeling translates complex, real-world operational challenges into rigorous quantitative frameworks. This responsibility includes identifying decision variables, defining objective functions, mapping physical and logical constraints, selecting appropriate mathematical structures (e.g., linear, non-linear, integer), and ensuring the model accurately reflects the underlying system dynamics.

Agentic skills contribute extensively to algebraic drafting, variable indexing, constraint generation, data structure mapping, and initial code scaffolding for mathematical programming libraries. Human skills remain essential for abstracting physical reality into mathematical logic, identifying hidden systemic constraints, determining the appropriate level of model granularity, and deciding which real-world complexities must be preserved versus simplified for computational tractability.

For example, an organization seeks to redesign its multi-tier service network to minimize response times while adhering to strict capacity limits. The operations research analyst clarifies the physical realities of the network, defines the boundaries of the system, and determines that a mixed-integer programming approach is required to handle facility location decisions. An AI agent then drafts the foundational algebraic equations, generates the index sets for the network nodes, and writes the initial constraint logic. The analyst reviews the formulation and identifies that the agent failed to include a critical non-linear constraint regarding minimum throughput thresholds for active facilities. After the analyst provides the corrective mathematical logic, the agent updates the formulation and verifies the dimensional consistency of the equations.

1.4.2. Optimization and Prescriptive Analytics

Optimization and prescriptive analytics solves mathematical models to identify the best possible decisions, resource allocations, or operational strategies. This responsibility includes configuring mathematical solvers, selecting appropriate algorithms (e.g., branch-and-bound, gradient descent), managing computational complexity, extracting dual variables and shadow prices, and evaluating tradeoffs between competing objectives.

Agentic skills are well suited to solver configuration, algorithmic code generation, automated parameter tuning, extraction of marginal values, and execution of multi-objective optimization routines. Human skills contribute to defining the relative weights of competing objectives, interpreting the economic or operational meaning of shadow prices, recognizing when a solver is trapped in a local optimum, and deciding whether a mathematically optimal solution is practically implementable.

For example, an analyst must determine the optimal production schedule for a complex manufacturing environment with severe changeover costs. The analyst defines the primary objective of minimizing waste while maintaining a secondary objective of maximizing on-time delivery. An AI agent generates the solver scripts, configures the branch-and-cut parameters, and executes the optimization. The agent returns a mathematically optimal schedule, but the analyst reviews the solution and recognizes that it requires an operationally impossible sequence of machine calibrations that the model approximated too loosely. The analyst instructs the agent to introduce a set of logical cutting planes to forbid these specific sequences, and the agent re-runs the solver to find a slightly sub-optimal but operationally viable schedule.

1.4.3. Simulation and Stochastic Modeling

Simulation and stochastic modeling evaluates system behavior under uncertainty, variability, and dynamic conditions where analytical optimization is intractable. This responsibility includes designing discrete-event or agent-based simulations, selecting appropriate probability distributions, modeling stochastic processes, generating random variates, and analyzing the resulting output distributions to understand system risk and variability.

Agentic skills assist with simulation scripting, random variate generation, automated replication execution, output logging, and basic statistical summarization of simulation runs. Human skills remain important for designing the logical flow of the simulated system, selecting distributions that accurately reflect real-world variance, identifying emergent behaviors, and determining whether the simulated scenarios adequately represent extreme but plausible operational shocks.

For example, a healthcare network wants to understand the impact of unpredictable patient surge events on emergency department wait times. The analyst defines the logical patient flow, identifies the key stochastic variables (arrival rates, triage times, treatment durations), and selects the appropriate probability distributions based on historical variance. An AI agent then writes the discrete-event simulation code, configures the Monte Carlo replications, and executes thousands of scenarios. The agent generates a summary of average wait times, but the analyst reviews the output distributions and notices the tail-end percentiles (representing catastrophic delays) are thinner than expected. The analyst directs the agent to introduce a specific correlated shock variable representing simultaneous staff shortages, re-runs the simulation, and evaluates the newly revealed systemic vulnerabilities.

1.4.4. Decision Analysis and Risk Evaluation

Decision analysis and risk evaluation structures complex choices under deep uncertainty, evaluating multiple criteria and long-term consequences. This responsibility includes constructing decision trees, applying multi-criteria decision analysis (MCDA), calculating expected utilities, evaluating the value of information, and framing risk tolerance levels for organizational leaders.

Agentic skills retrieve decision-analytic frameworks, generate decision tree structures, calculate expected monetary values, perform automated utility calculations, and draft comparative summaries of strategic alternatives. Human skills evaluate the organization’s actual risk appetite, define the qualitative criteria that must be included in the MCDA, assess the ethical or strategic implications of different decision branches, and determine whether the mathematical utility functions accurately capture executive preferences.

For example, an executive team must decide whether to invest heavily in a new, unproven technology or incrementally upgrade existing infrastructure. The analyst defines the decision criteria, including financial return, strategic alignment, and regulatory risk. An AI agent constructs a complex decision tree, populates it with probabilistic market forecasts, and calculates the expected monetary value for each branch. The analyst reviews the model and identifies that the agent’s utility function assumes a strictly risk-neutral posture, which contradicts the board’s current mandate to preserve capital. The analyst revises the utility parameters to reflect risk-averse preferences, instructs the agent to recalculate the certainty equivalents, and uses the adjusted model to recommend a more conservative, phased investment strategy.

1.4.5. Model Validation, Sensitivity Analysis, and Testing

Model validation, sensitivity analysis, and testing ensures that mathematical and simulation models are robust, accurate, and reliable before they are used to inform consequential decisions. This responsibility includes verifying code logic, validating outputs against historical data, performing automated parameter sweeping, extracting sensitivity ranges, and stress-testing models against edge cases.

Agentic skills contribute to automated unit testing, systematic parameter variation, shadow price extraction, historical backtesting routines, and the generation of tornado diagrams or spider plots. Human skills define the acceptable boundaries of model error, interpret the operational significance of sensitivity ranges, identify structural flaws that cause unrealistic behaviors, and determine whether the model is sufficiently validated for deployment.

For example, an analyst develops a sophisticated network optimization model for global freight routing. Before presenting the model to stakeholders, the analyst defines the critical cost parameters that require rigorous stress testing. An AI agent executes an automated sensitivity analysis, systematically varying fuel costs, tariff rates, and transit times across thousands of combinations, and generates a comprehensive dashboard of shadow prices and reduced costs. The analyst reviews the sensitivity dashboard and identifies that the model’s optimal routing is hyper-sensitive to a specific regional tariff that is currently under political negotiation. Recognizing this vulnerability, the analyst instructs the agent to formulate a robust optimization variant that hedges against this specific parameter’s variance, ensuring the final recommendation remains stable under multiple future scenarios.

1.4.6. Operational Implementation and Decision Support

Operational implementation and decision support translates complex mathematical results, algorithmic outputs, and simulation findings into actionable business strategies and operational rules. This responsibility includes explaining prescriptive recommendations, visualizing optimization tradeoffs, integrating model outputs into operational software, managing change resistance, and monitoring the real-world performance of implemented solutions.

Agentic skills summarize algorithmic outputs, generate executive dashboards, draft implementation documentation, format data feeds for downstream operational systems, and track post-implementation performance metrics. Human skills translate mathematical jargon into operational business rules, manage stakeholder expectations, negotiate the practical compromises required to deploy a theoretically optimal solution, and determine when a model’s recommendations must be overridden by human operational expertise.

For example, an optimization model recommends a radical redesign of a company’s regional inventory placement, which theoretically saves millions in holding costs. The analyst reviews the output and recognizes that the model assumes instantaneous lateral transshipments between warehouses, which the warehouse managers know is operationally chaotic during peak seasons. The analyst translates the mathematical shadow prices into a set of practical “business rules” and heuristic buffers that the warehouse managers can actually execute. An AI agent then formats these adjusted rules into the enterprise resource planning (ERP) system’s configuration files and generates a visual dashboard comparing the theoretical optimum with the practically adjusted implementation. The analyst presents this compromise to the executive team, securing their approval for the operationally viable solution.

1.5. Changing Human and Agentic Involvement Across Operations Research Responsibilities

Human skills and agentic skills are involved throughout operations research, but their relative involvement differs across responsibilities and changes during task execution.

These involvement levels are not permanent allocations. A single responsibility may contain actions where human and agentic skills compete as alternative capabilities and other actions where they collaborate sequentially, iteratively, or recursively.

The practical objective is to construct skill configurations that use the most suitable available capabilities for each part of the work and revise those configurations when task performance evaluation shows that another arrangement performs more effectively or reliably.

1.6. Working as an Operations Research Analyst with Agentic AI

Agentic AI is becoming an increasingly powerful component of operations research. Operations research analysts should therefore adopt and integrate AI agents as complementary participants in professional workflows rather than treating them only as occasional tools for generating code, summarizing algorithms, or formatting sensitivity tables.

This shift does not reduce the importance of mathematical rigor and operational expertise. Human professionals remain responsible for the occupation and its corresponding responsibilities. They understand systemic constraints, define meaningful objective functions, interpret physical realities, determine appropriate modeling methodologies, evaluate algorithmic tradeoffs, communicate with operational stakeholders, control consequential implementation decisions, evaluate task performance, and decide how human and agentic skills are configured and improved.

At the same time, agentic skills greatly expand the volume and complexity of mathematical work that can be processed. AI agents retrieve and organize algorithmic literature, generate and debug complex solver scripts, configure optimization parameters, execute thousands of stochastic simulations, perform automated sensitivity sweeps, draft implementation documentation, and synthesize extensive prescriptive evidence at a scale difficult to achieve through direct human effort alone.

Strong operations research practice is therefore not based on maximizing human work or agentic work independently. It is based on finding appropriate configurations of human skills and agentic skills for the responsibilities, tasks, subtasks, and actions being performed.

Where a human skill and an agentic skill can independently perform the same action, skills competition identifies which is better suited to the current context. Where work requires complementary capabilities, skills collaboration results in human and agentic skills being activated sequentially, iteratively, or recursively to achieve stronger task performance than either skill type achieves alone.

Task performance evaluation then determines whether selected skills and configurations satisfy predefined expectations. Successful skills are retained and reused, while inadequate skills are refined, reconfigured, or newly developed. The resulting capabilities return to an increasingly useful skill space that supports operations research and, where capabilities are reusable, other occupations as well.

For an operations research analyst, working effectively with agentic AI means treating AI as a powerful complementary capability integrated into everyday professional routines while maintaining human control of occupational responsibilities. Through appropriate skill selection, competition, collaboration, evaluation, and continuous improvement, human professionals and agentic AI work together to achieve operations research that is more efficient, mathematically rigorous, computationally tractable, operationally viable, and fundamentally aligned with organizational strategy.