What's New in Optilogic? Product Update: October 2026

Great supply chain decisions start with models that look like your business. That's why this month, every model gets closer to the way your operations run. The result: answers you can take to leadership, backed by a plan your team can execute.

Ada now works with the files and workflows you already have, and reasons across inventory, risk, and simulation. Cosmic Frog now models what teams used to approximate: dock door limits, rush hour traffic, multi-level bills of materials, and warehouse labor. DataStar lets you clean Excel data once, before it ever enters a workflow.

Together, these updates help you optimize the supply chain you have today and design the one you need for tomorrow, with every output open for your team to check.

Ada

Attach Files Directly to a Conversation

You no longer need to set up a data connection and import your data before Ada can see it. Attach a file and start asking questions in seconds.

Ada accepts these file types:

• CSV and Excel (.xlsx, .xls)
• PDF
• Text (.txt) and Markdown (.md)
• JSON

There are three ways to attach. Click the plus icon below the prompt box, then choose:

1. Upload from computer to select one or more local files. ‍
2. Upload a folder to attach every file in a folder and its subfolders. ‍
3. Attach workspace file to point Ada at files already in your Explorer workspace.

Once a file is attached, here is what you can ask Ada to do with it:

• Spot-check data quality before it goes anywhere near a model. ‍
• Summarize a document and answer questions about its contents. ‍
• Analyze the data directly in the conversation. ‍
• Load it into a Cosmic Frog model or DataStar project sandbox so it is ready for modeling.

Learn how to attach files in Getting Started with Ada & Agentic AI

Ada Now Covers Inventory, Risk, and Simulation-Optimization

Ada now supports three more Cosmic Frog engines. Anyone doing safety stock, risk, or reorder point work has a reason to bring Ada into that workflow, not just network design.

‍

Where to build, how to move freight, where to hold inventory, where the risk sits, and whether the plan survives: one assistant, the same data.

Ada and DataStar: Understand, Fix, and Build Your Macros

Ada now works more seamlessly with your DataStar projects. It reads your macros, each step and how it's configured, and your recent run history. Its answers come from your actual project, not a guess.

Start with a problem.  

Try asking: "My Clean Raw Data macro failed last night. What went wrong, and can you fix it?"

Ada checks the run history and finds the step that failed. For example, a ship date column may have come through blank because the source file mixed two date formats. Ada explains the cause in plain English, updates the step to handle both formats, and checks the macro before you rerun it. You don't have to dig through logs.

Make sense of what you inherited.  

For macros someone else built, or that you built months ago, just ask:

"Summarize what each macro in this project does."

"Explain how transportation policies are created in the Build Policies macro."

"How often has each macro run, and how often did it fail?"

Build by describing what you need.  

Ada can add a single task or build a full macro:

One task: "Add a step that removes duplicate customer records."

A full macro: "Build a macro that imports the shipments file, standardizes postal codes, and totals volume by lane."

Built-in guardrails. Ada tells you which project it's about to change and checks for problems before a run. It flags a failure only when there really is one.

The result is less time reverse-engineering macros and chasing false alarms, and more time on the analysis. It's one more way Ada shows its work.

See more example prompts in DataStar Quick Start: Macro Building with Ada

Hopper

Two Hopper updates this month share one goal: a routing plan your dispatch team can execute on Monday morning, not one that only works on paper.

Dock Door Constraints

Site congestion is one of the most common reasons a plan turns out not to be executable. A depot with two dock doors can only service two trucks at once. Until now, Hopper assumed any number of vehicles could load or unload at the same time, so teams approximated the limit with time windows.

Now you can model it directly. Questions you can answer:

• Are my planned routes feasible given the dock doors at my depots and customers?
• Where are dock doors the bottleneck, and what do they cost me in wait time or extra routes?
• How many doors does a depot need to avoid delays?
• What happens when a depot runs with fewer doors during maintenance or construction?

How it works. Set the new Number of Dock Doors field on the Facilities or Customers table.

‍

Only loading and unloading occupy a door. A truck that is waiting does not. Make sure pickup and delivery times are set, since those determine how long each truck holds a door.

When a dock door conflict comes up, Hopper works through it in order:

1. Reschedules tours and shifts stop timing.
2. Moves stops, tries shipments on other routes, or adjusts the asset mix.
3. Drops a tour only when rescheduling cannot resolve the conflict, and flags those shipments as Unrouted.

To illustrate the difference this makes, the Help Center walks through an example built on the Multi Stop Routes with Time Windows template model in the Resource Library. Limiting an Atlanta depot to a single dock door left cost, distance, and drive time unchanged. Only one route's start time moved, turning a schedule that worked on paper into one the depot could run.

Read Dock Door Constraints (Transportation Optimization)

Time-of-Day Transit Times

A single average transit time over- or under-states the plan for urban and peak-sensitive fleets. Now a lane can carry different distance, time, and cost depending on when the vehicle departs, such as:

• Rush hour
• Overnight
• Weekends

How it works. A new TransitMatrixSchedule column on the TransitMatrix table references a BusinessHours schedule. Hopper then shifts tour start times to take advantage of the faster windows.

Cyclo

Bills of Materials for Multi-Level Production

Manufacturers can now size safety stock at every stage of a multi-level product structure in a single model, not just at finished goods.

Model the full structure, from components to sub-assemblies to end products, using the BillsOfMaterials and ProductionPolicies tables. Cyclo then:

1. Propagates demand through the bill of materials.
2. Handles make-and-buy decisions.
3. Places safety stock at component stages.
4. Flags any BOM cycles with clear errors.

Get started with Cyclo

Neo

Faster Runs, Lighter Footprint

A series of improvements makes Neo materially faster and lighter. We measured the results across a benchmark suite of 28 customer model and scenario pairs.

‍

What this means for your work:

• Smaller resource sizes. Large models that needed XL or 2XL resources should now fit on smaller ones. ‍
• Build time steps back. Model build no longer dominates total runtime. ‍
• Fewer failed runs. Models that used to run out of memory while writing results are far more likely to complete.

Review Resource Size Selection Guidance

Warehouse Handling, Now Part of the Optimization

Warehouse handling work is now part of the optimization, so labor and equipment capacity limits and their costs show up directly in your results.

Attach a stocking or destocking process to a facility and product's inbound or outbound flow. It consumes Work Center capacity just like a production process does.

Where to find the results:

‍

Brush up on Work Centers and Processes in Detailed Production Modelling

DataStar

Excel Connection Improvements: Clean It Once, at the Source

Enhanced Excel Connections let you configure and clean workbook data before it ever loads into DataStar. You can:

• Select which sheets to load
• Define headers
• Choose and rename columns
• Override data types
• Manage blanks and nulls
• Set decimal formats

All of it lives at the connection level.

That means you're not editing the source file, and you're not repeating the same fixes in every downstream macro that references it. Clean it once. It stays clean everywhere that connection is used.

Put October's Releases to Work

Want help applying these updates to your own models? Your Customer Success Manager and our Solutions team can walk through dock door constraints, BOM-level safety stock, or Ada workflows with your data.

All features are available now in the Optilogic platform. Have questions? Reach out to your Customer Success Manager, email support@optilogic.com, or visit our Help Center for documentation and tutorials. New to Optilogic? Watch the five-minute demo.

‍