Skip to main content

​
Overview

ProDex’s AI assistant is called Dexter — an AI agent built into the platform with direct access to your factory data. Dexter can ingest and transform your operational data, build simulation models, run analyses, generate reports, and make changes to your configuration, all through natural language conversation. This isn’t a chatbot that answers general questions. Dexter is an operational tool that works with your actual data. When you ask it to “show me the bottleneck in Line 3,” it writes a query against the simulation’s raw event data, reads the results, and tells you — with numbers. When you upload a messy MES export, it profiles the data, asks you to clarify what the status codes mean before assuming, and traces every derived value back to its source. Dexter defaults to asking rather than guessing. At major decision points, it presents structured questions with clear options; when interpreting your data, it flags anything it’s uncertain about — ambiguous column names, unlabeled units, codes without a legend — and asks you to confirm before building anything.

​
What Dexter Can Do

​
Ingest and Transform Data

Upload production logs, MES exports, inventory snapshots, or any operational data. Dexter will profile the file — columns, data types, value distributions, row counts — and ask you to clarify anything ambiguous before proceeding. When the work goes beyond simple inspection, Dexter creates a documented data pipeline that traces every derived value back to its source data: what file it came from, what query produced it, what assumptions were made, and how confident the result is. This means you can always audit how a number was calculated. And when you upload fresh data next week, the same pipeline can be re-executed to produce updated outputs without rebuilding anything.

​
Analyze and Explain

  • Interpret simulation results — throughput, utilization, cycle times, bottlenecks
  • Compare runs side by side and explain what changed
  • Run Monte Carlo analysis — hundreds of simulation replications to quantify variability, build confidence intervals, and identify which metrics are stable vs. sensitive to randomness
  • Explore your bill of materials — explode a BOM to see its full component tree, or do a reverse lookup to find everywhere a component is used
  • Answer questions about your model’s structure and configuration
  • Summarize experiment results and highlight key findings

​
Build and Modify

  • Create and edit model components — sources, processes, resources, buffers, routers, combiners, separators, transformers, and connections
  • Define KPIs and charts for your results dashboard — each backed by a precise, auditable query against the simulation data
  • Set up constants, lookup tables, and entity definitions
  • Configure what-if experiments with multiple scenarios for side-by-side comparison
  • Build schedules that define material releases, shift patterns, and capacity changes

​
Run and Test

  • Execute simulations and wait for results
  • Run what-if experiments across multiple scenarios with structured KPI comparisons
  • Run Monte Carlo batches for statistical analysis across many replications
  • Test parameter changes and report the impact

​
Generate Reports

  • PDF reports — polished, branded documents with KPI summaries, charts, and narrative analysis. Suitable for sharing with stakeholders who don’t use ProDex directly.
  • Excel exports — structured data tables for further analysis, filtering, and pivoting in your own spreadsheets.
  • Word documents — editable narrative reports you can refine, reformat, or drop into a larger document.
  • PowerPoint decks — slide-formatted summaries for meetings and executive reviews.
Reports are generated on demand — just ask. For example: “Generate a PDF summary of the last run,” “Export the throughput data to Excel,” or “Build me a deck for Monday’s review.”

​
Automate with Scheduled Jobs

Dexter can run tasks on a schedule without manual intervention. Set up recurring jobs to:
  • Run a simulation every Monday morning and generate a report with the results
  • Produce a weekly PDF summarizing the latest metrics
  • Execute a standard analysis workflow on a recurring cadence
  • Re-run a data pipeline when fresh data is uploaded on a regular cycle
Scheduled jobs use cron expressions for flexible timing — daily, weekly, specific days and times, or custom intervals. Each job runs a full conversation with Dexter, so it has access to all the same capabilities you’d have interactively.

​
Starting a Conversation

Open the Dexter panel from the sidebar. You can have multiple conversations open — each one maintains its own context and history. Start by describing what you want in plain language. You don’t need to use specific commands or syntax. Some examples:
  • “What’s the throughput of my current model?”
  • “Add a buffer between the assembly process and the paint station with capacity 50”
  • “Run the simulation and show me which resources are over 90% utilization”
  • “Compare the last two runs and tell me what improved”
  • “Upload this MES export and derive cycle times by station”
  • “Run 100 replications and tell me how confident we can be in the throughput number”
Dexter will ask clarifying questions if it needs more information. It uses a structured question format with clear options, so you can respond with a click rather than typing.

​
How It Works

When you send a message, Dexter reads your current factory context — your model, entities, constants, schedules, and recent results. It can then take actions by reading files, writing configurations, and executing commands in a secure sandbox environment. Every action Dexter takes is visible in the conversation. You’ll see tool calls (file reads, writes, command executions) inline so you can follow exactly what it’s doing and why.

​
File Attachments

You can attach files to your messages — PDFs, CSVs, Excel spreadsheets, PowerPoint decks, Word documents, or images. Dexter handles both structured data (CSVs, spreadsheets) and unstructured content (PDFs, decks, Word docs, images), so you can share operational knowledge whether it lives in clean tables or narrative documentation. Common uses:
  • Uploading production data for analysis or model parameterization
  • Sharing reference documents with operational context — SOPs, equipment manuals, team decks
  • Providing spreadsheets to import into your model
  • Uploading order documents for BOM configuration
  • Capturing data from dashboards or systems without a direct integration — screenshot a BI view or export a report, and Dexter can work with it
Uploaded files persist across conversations, so you can reference them later. You can manage your uploaded files through Settings in the top right corner of the chat window, under AI Assistant settings.

​
Connected Data Sources

Beyond files you attach to messages, Dexter can also query your connected data sources — warehouses, databases, ERPs, and MES systems your ProDex team has linked to your factory. Unlike file attachments (which are per-message uploads), connected sources are persistent: Dexter can query them any time, in any conversation, against live data. See Data Sources for the list of supported systems and how connections are set up.

​
Persistent Knowledge Base

Dexter builds a cumulative understanding of your operation across every conversation. When you explain that “Code 3 means rework” or “we run two shifts with a skeleton crew on weekends” or “we call that station the bottleneck cell,” that knowledge is stored and applied in future sessions — you never have to repeat yourself. Over time, this creates a structured record of the operational knowledge that traditionally lives only in your most experienced team members’ heads: process rules, naming conventions, data quirks, KPI targets, shift patterns, and decision logic. This isn’t chatbot memory — it’s a persistent, organized knowledge base scoped to your factory that grows more valuable with every conversation.

​
Workflows

Workflows are reusable task patterns that guide Dexter through multi-step operations. They ensure consistency and completeness for common tasks. ProDex includes built-in workflows for common operations. Your team can also create custom workflows tailored to your specific processes — codifying your best practices into executable procedures that Dexter follows consistently. When a process that used to take a senior engineer an hour to explain becomes a workflow, any team member can trigger it and get the same reliable result.

​
Best Practices

​
What It Does Well

  • Operational analysis with your actual data
  • Ingesting messy real-world data and transforming it into structured, traceable model inputs
  • Multi-step tasks that combine building, running, and analyzing
  • Generating formatted reports for stakeholders
  • Answering “what if” questions by running structured scenario comparisons
  • Automating recurring analysis with scheduled jobs
  • Building a persistent knowledge base of your operation over time

​
Tips for Best Results

  • Be specific about what you want. “Show me utilization” is good. “Show me utilization for the CNC machines over the last 5 runs” is better.
  • Let Dexter ask questions. When Dexter asks clarifying questions, it’s being thorough — not slow. Responding to these produces better, more accurate results than trying to specify everything upfront.
  • Review changes before running. When Dexter modifies your model, review the changes in the conversation log before running a simulation.
  • Use it iteratively. Start with a question, review the answer, then refine. Dexter maintains full context within a conversation, so follow-up questions build on what came before.
  • Teach Dexter once. When you correct it or explain something about your operation, Dexter saves it to the knowledge base. Next conversation, it should apply what it learned — if it doesn’t, flag it, and that becomes the new baseline.