Decision Workflows

This page maps common user questions to the right OpenPinch entry point, example asset, and supporting documentation page.

Which Workflow Should I Use?

Can I run a known-good case and inspect the main targets?

Use basic_pinch.json with First Solve with Python. If you want the packaged learning path first, use Working with the CLI to copy notebook 01, then solve inside Python from that notebook.

How sensitive is the answer to my minimum approach assumptions?

Use crude_preheat_train.json and 01_basic_pinch_and_dtcont_sensitivity.ipynb. The notebook uses PinchWorkspace.copy_case(...) and set_dt_cont_multiplier(...) to keep the comparison explicit.

How do multiple process areas aggregate into a site view?

Use zonal_site.json or pulp_mill.json together with Zonal and Total Site Workflows and 02_total_site_targets_and_sugcc.ipynb.

Would an integrated heat pump improve the utility picture of my plant?

Use heat_pump_targeting.json with Heat Pump Workflows. The dedicated explicit problem.target.direct_heat_pump(...) / problem.target.indirect_heat_pump(...) workflows are the supported advanced route.

How do I compare direct and indirect HPR or refrigeration targets?

Use chocolate_factory.json and 03_carnot_hpr_comparison.ipynb. That workflow stays on the public problem.target.* and problem.plot.* surfaces rather than lower-level cycle internals.

I need a typed request/response service contract, not a notebook wrapper.

Start from Service Layer and Schemas and Config.

I need to inspect prepared streams, zones, or Problem Tables directly.

Start from Domain Model.

Interpretation Sequence

Regardless of workflow, the recommended decision sequence is:

1. compare the hot and cold utility targets first

2. compare heat recovery and pinch temperatures second

3. inspect the most relevant graph family third

4. only then move into advanced study-case or equipment interpretation

That order keeps the package grounded in thermodynamic decision support rather than graph-first exploration.