Overview
ProDex is an AI-native platform for manufacturing operations. It combines discrete event simulation — modeling your operation as a stream of timestamped events — with data pipelines, production planning, and job scheduling in a single environment. Running through all of it is Dexter, an AI agent that can build models, run analyses, generate reports, and automate workflows on your behalf. Everything you build lives in a Factory — a single workspace container that holds your simulation models, data pipelines, BOMs, plans, schedules, experiments, insights, and run history. Your whole team works against the same Factory. The core idea: instead of managing your operations across spreadsheets, point tools, and manual reports, ProDex gives you a live model of your factory that your team and AI can both work with directly.Who Is It For?
ProDex is built for the people who run manufacturing operations day-to-day:- Operations managers who need to understand throughput, bottlenecks, and capacity
- Production planners who translate demand into achievable production schedules
- Schedulers who assign jobs to machines and workers
- Analysts who want to model what-if scenarios before making decisions
Where to Start
Most new users follow this path:- Bring your data in — upload spreadsheets, MES exports, or PDFs through Pipelines, where Dexter profiles your data and turns it into reusable, auditable inputs.
- Build a model — describe your operation to Dexter or drag components onto the Modeler canvas to get a flow graph of your factory.
- Run simulations — see throughput, bottlenecks, and resource utilization with full event traces.
- Compare scenarios — use Experiments to lay variants out side by side and quantify the difference.
- Decide — generate plans, schedules, and reports backed by what your model actually predicts.
Key Capabilities
Modeler
Build a visual model of your factory as a flow graph — sources, processes, stations, buffers, routers, and sinks connected by paths that entities (parts, batches, work orders) follow through your operation. Assign shared resources like operators and machines, and use combiners, separators, and transformers to model assembly, splitting, and conversion. Run discrete event simulations to understand how your operations perform over time, where bottlenecks form, and how changes ripple through the system. Schedules anchor your simulation to real-world timing — define when material enters the system and trigger timed actions during the run. Resources define their own availability patterns, which is how you model shifts and capacity changes. Each execution produces a detailed timeline of what happened, where, and when — so you can validate a plan before committing to it on the floor. For statistical analysis (confidence intervals, percentile distributions, variability profiling), wrap any run in Monte Carlo — same model, many random seeds.Pipelines
Pipelines are how raw operational data becomes platform inputs. Upload an MES export, an ERP report, even a screenshot of a dashboard — Dexter profiles it, transforms it, and writes a reusable, auditable pipeline that produces concrete derivations (cycle times, processing times, demand profiles) you can wire directly into your simulation, planning, or BOM models. Every value is traceable back to its source.Experiments
Snapshot model variants and compare them side by side — different schedules, different resource configurations, different demand profiles. Experiments lay KPIs and charts out across snapshots so you can quantify the difference, not just describe it. When the question is “should we add a CNC?” or “what happens with the holiday schedule?”, this is the surface that answers it.Insights
Free-form dashboards for exploratory data analysis. Build chart panels from pipeline outputs or uploaded files when you want to look at your data visually before or alongside building a model. Insights live at factory scope, so anyone with access sees the same set.Production Planning
Given demand orders, current inventory, and your bill of materials, ProDex’s planning optimizer outputs a production plan — what to make, when, and in what quantities across planning periods. The optimizer balances demand fulfillment against inventory goals and resource capacity constraints to find the best feasible plan, and supports custom constraints for production smoothing, minimum lot sizes, and other business rules.Job Scheduling Beta
Go from plan to floor-level execution. Given a set of production orders, the job scheduler assigns specific jobs to specific resources and sequences them — accounting for processing times, changeover costs, and resource constraints to produce an optimized schedule. Combined with simulation, you can validate the schedule against real-world variability before committing to it.Dexter
ProDex’s AI agent — an operations research specialist with direct access to your factory model, your data, and your results. Describe your operation in plain language and Dexter will build a simulation model, analyze results, and generate reports. Upload production data and Dexter will profile it, derive model parameters, and document every transformation so you can trace any number back to its source. Dexter works with you iteratively — asking questions to verify assumptions and showing its work at each step rather than guessing in silence. It also maintains a persistent knowledge base of your operation across conversations — your processes, your terminology, your business rules — so you don’t have to re-explain your operation every time you come back. The more you work with it, the less you have to repeat yourself.Module Summary
| Module | What it does | Key question it answers |
|---|---|---|
| Modeler | Model your factory as a flow graph and run discrete event simulations | ”What throughput can we actually achieve?” |
| Pipelines | Transform uploaded data into auditable platform inputs | ”How do I get my MES/ERP data in?” |
| Experiments | Compare model variants side by side | ”What happens if we change X?” |
| Insights | Free-form dashboards for exploratory analysis | ”What does my data actually look like?” |
| Production Planning | Optimize what to produce and when across periods | ”What should we make, when, and how much?” |
| Job Scheduling | Assign jobs to resources at the floor level | ”Who does what, on which machine, in what order?” |
| BOMs | Define product recipes and variants | ”What goes into building each product?” |
| Configurator | Define configurable product structures and option classes | ”How do my product options translate into what gets built?” |