A real control-logic language
Per-step logic with set, increment, conditionals, loops, waits, get and free, and signals, with arrays and attributes, edited inline with highlighting and suggestions.
Technical capabilities
Built for complex operational systems
The question for any serious evaluation is simple. Can the tool represent the real complexity of your operation, and can it analyze that model fast enough to support a decision? ProcessModel is built to answer yes to both.
AI-assisted model development
AI where it is strongest. Simulation where it is strongest.
AI gets you to a working model quickly and helps you read the results. A real simulation does the rigorous analysis that a decision rests on. Each does what it is genuinely good at.
Where AI is strongest
- Interpret a process description in plain language
- Build the model structure for you
- Suggest logic the model is likely missing
- Identify the constraints worth a closer look
- Explain the run results and what they mean
Where simulation is strongest
- Test scenarios repeatedly under real conditions
- Handle variability, queues, resources, and routing
- Respect schedules, capacities, and costs
- Run many replications for stable numbers
- Compare alternatives with measurable results
Visual modeling
Model the process as it actually works
A process is a network of activities, queues, resources, routes, and decisions. You build it that way on the canvas, connecting the steps and the paths so the model reads like the operation it stands for.
As a model grows into dozens of connected steps, one-click automatic layout untangles it for you, so a complex network stays readable instead of turning into a knot of crossing lines.
- Activities, queues, resources, routes, and decisions on one canvas.
- One-click automatic layout keeps a large model readable.
- Collapsible sub-models keep the bigger picture in view.
Rapid experimentation
Test the future state before you build it
A scenario is a set of changes layered over the base model. Adjust resources, schedules, routing, batch sizes, capacities, priorities, and decision logic, then run it and see what happens.
Keep the current state intact and compare it against the future state with measurable results, so every proposed change is backed by numbers rather than opinion.
- Override resources, schedules, routing, capacities, and logic.
- Compare current versus future state with real numbers.
Performance
Analyze a complex model fast
Rigorous analysis is only useful if the answer arrives in time to act on it. ProcessModel runs replications and experiments in parallel, so a model with real variability still returns stable numbers quickly.
Parallel replications
Many replications of the same model run at once, so the variability settles into reliable averages without a long wait.
Parallel experiments
Several scenarios run side by side, so a sweep of alternatives is measured together rather than one slow run after another.
Stable numbers, sooner
High-volume models with real arrivals and queues return confidence intervals across replications fast enough to keep a review moving.
What it can represent
Twelve domains of operational detail
The building blocks for a model that behaves like the real operation, from flow and variability to control logic and cost.
Flow and routing
Percentage, conditional, fork, pull, load balancing, attach and detach, renege, and fallback paths that mirror how work really moves.
Time and variability
26 probability distributions, plus distributions fitted from your own observed data, seeded streams for reproducibility, and warm-up periods.
Demand and arrivals
Continuous, periodic, scheduled calendar grids, daily patterns with a 24-hour curve, and order lead time, with bulk import for large arrival files.
Resources and labor
Pooled staff and machines, availability and breakdowns, ten shift presets and custom schedules, breaks, overtime, priorities, and travel time.
Queues and blocking
Finite or unlimited queues, input and output buffers, back-pressure and blocking, and disciplines from FIFO to earliest due date.
Batching and material handling
Batch by count, timeout, schedule, or rule, with selection rules, attribute aggregation, nested batching, unbatch math, and carriers with finite capacity.
Inventory
Aggregated stock, multiple items, zones and locations, unit costs, storage capacity, reorder levels, substitution, and FIFO or LIFO consumption.
Control logic
Gates and signals plus a per-step discrete-event language with variables, attributes, and one, two, and three dimensional arrays, and decision tables.
Cost and value stream
Resource, activity, entity, and per-move costs, with value-added, required non-value-added, and non-value-added time tracked across every path.
Entity state
Type, custom attributes, carrying capacity, parent and child links, age, time in system, accumulated time and cost, and live status, all queryable in expressions.
Scale and hierarchy
Collapsible sub-models and process groups, callable subprocesses, asynchronous exit, input and output mapping, and high-volume capacity.
Calendars
Weekly shift schedules with presets and custom, breaks, multi-shift and around-the-clock, time-of-day and day-of-week logic, and elapsed or wall-clock time.
Variability fitted to your data
Choose from 26 distributions, or paste your own observed data and let the tool rank the best fit, so timing reflects reality instead of a convenient guess.
Results and analysis
Answers, not just a finished run
Every run produces a full report. The cost and value-stream view shown here breaks down where money goes by resource, activity, and entity, alongside the rest of the model in its own tab.
The Output Report has a tab for every part of the model
Twelve tabs cover the summary, activities, entities, gates, process tables, resources, routes, variables, arrays, costs, impact, and queues, with confidence intervals across replications (90%, 95%, or 99%).
Impact Analysis ranks the real constraint
Every step is ranked by the time and cost it adds, so the constraint that matters stops hiding behind the obvious one.
AI Analysis reads the results with you
It explains the run in plain language, then answers your follow-up questions with suggested prompts, grounded in your own activity, resource, and entity names. Set a focus first: throughput time, quantity, cost, process efficiency, or resource utilization.
Queue Diagnostics shows when work builds
A distinctive view of where and when queues form through the operating window, with the worst hours called out.
Cost and value-stream view
Cost by resource, activity, and entity, with value-added and non-value-added time, charted and exportable.
Proof
Trusted on the hard models
Founded in 1999, ProcessModel has helped Fortune 500 companies save billions across manufacturing, healthcare, consulting, and government.
Our people thought they would have to hire and train an additional 500 representatives. The modeling pointed out where the real problems were, and how to re-engineer the process with the addition of only 125. That is a 14 million dollar saving.
The most easy to learn, most powerful, and most useful simulation and modeling software for transactional processes.
Represent the complexity. Analyze it fast.
Build the network, model the variability, write the logic, run it in parallel, and read the results. The detail you need to trust the answer, with the speed to act on it.