[Econ] Preprocess Seed Costs by allisterakun · Pull Request #3112 · RuminantFarmSystems/RuFaS

allisterakun · 2026-06-24T17:07:44Z

Adds seed cost preprocessing to the Economics module, computing per-field seed costs by mapping crop specifications to commodity prices and converting field sizes to m².

Context

Issue(s) closed by this pull request: closes #3080.

What

Added seed cost preprocessing logic in preprocessing.py that reads field.crop_specification and field.field_size from the simulation data, converts field size to m² (to match economics inputs), and maps crops grown to the corresponding commodity prices to compute seed costs per field.
Added "seed_cost": [0.0] to the fallback values dictionary in fallback_values.py.
Added a print statement in EEE_manager.py for the metadata load failure case (debugging aid).
Updated seed cost reporting per Braden's feedback.
Added unit tests in test_economics_preprocessing.py for the new seed cost preprocessing logic.
Updated test_economics.py to account for the new seed cost data flow.

Why

The Economics module needed to process seed costs as part of the farm economic analysis, but there was no preprocessing step to extract crop specification and field size data from the biophysical simulation outputs, convert units, and map crops to their seed cost commodity prices. Without this preprocessing, seed costs could not be included in the economic analysis.

How

The preprocessing step reads field.crop_specification (which crops are grown on each field) and field.field_size (the area of each field) from the simulation data. Field sizes are converted to m² to match the units expected by the economics inputs. Crops are mapped many-to-one to commodity price categories using the existing crop-to-commodity mapping. The resulting seed cost per field is computed and made available to the economic framework. When biophysical data is unavailable, the fallback value of 0.0 is used for seed cost.

Test plan

Added unit tests in test_economics_preprocessing.py covering seed cost preprocessing with valid crop/field data, missing data fallback, and crop-to-commodity mapping.
Updated test_economics.py to include seed cost in the economics data flow.

Input Changes

N/A (uses existing field.crop_specification and field.field_size data from the biophysical simulation)

Output Changes

N/A

Filter

bradenlimb · 2026-07-20T00:40:09Z

Hey Allister

Looking at the inputs/outputs I have one question on the field size. The biophysical aggregate output is 400,000 m2 are planted with corn. However, in the inputs I only see corn_silage being planted twice in 2013/2017 and the field size is 10. I assume this field size is in hectares (please verify) which would mean a size of 100,000 m2 so the corn planted size should be 200,000 m2

Similar but different issues with alfalfa hay vs silage. Both are reported as 172,706 m2 by the code, but I only see alfafa_silage between planted twice so should have the same area as corn and alfalfa_hay should be zero.

I might be misfollowing the input json so please correct me if this is the case

allisterakun · 2026-07-20T14:21:18Z

Hey Allister

Looking at the inputs/outputs I have one question on the field size. The biophysical aggregate output is 400,000 m2 are planted with corn. However, in the inputs I only see corn_silage being planted twice in 2013/2017 and the field size is 10. I assume this field size is in hectares (please verify) which would mean a size of 100,000 m2 so the corn planted size should be 200,000 m2

Similar but different issues with alfalfa hay vs silage. Both are reported as 172,706 m2 by the code, but I only see alfafa_silage between planted twice so should have the same area as corn and alfalfa_hay should be zero.

I might be misfollowing the input json so please correct me if this is the case

I believe this is because there are two fields in this simulation. Both have a field size of 10 hectares.
Field 1 plants corn on 2013:127 and harvests on 2013:298, then again plants on 2017:127 with harvest on 2017:298.
Field 2 plants corn on 2013:127 and harvests on 2013:263, then again plants on 2017:127 with harvest on 2017:263.
That's why, when summing the total field size for corn, it comes out double what you expected.

Regarding alfalfa silage and hay: both are expected to be planted (silage in field 1 and hay in field 2). I'll look into the values for their field size!

bradenlimb · 2026-07-20T19:26:41Z

Hey Allister
Looking at the inputs/outputs I have one question on the field size. The biophysical aggregate output is 400,000 m2 are planted with corn. However, in the inputs I only see corn_silage being planted twice in 2013/2017 and the field size is 10. I assume this field size is in hectares (please verify) which would mean a size of 100,000 m2 so the corn planted size should be 200,000 m2
Similar but different issues with alfalfa hay vs silage. Both are reported as 172,706 m2 by the code, but I only see alfafa_silage between planted twice so should have the same area as corn and alfalfa_hay should be zero.
I might be misfollowing the input json so please correct me if this is the case

I believe this is because there are two fields in this simulation. Both have a field size of 10 hectares. Field 1 plants corn on 2013:127 and harvests on 2013:298, then again plants on 2017:127 with harvest on 2017:298. Field 2 plants corn on 2013:127 and harvests on 2013:263, then again plants on 2017:127 with harvest on 2017:263. That's why, when summing the total field size for corn, it comes out double what you expected.

Regarding alfalfa silage and hay: both are expected to be planted (silage in field 1 and hay in field 2). I'll look into the values for their field size!

Hey Allister

You are correct. I missed that there were two fields. Corn grain is correct at 400,000 m2. As far as I can tell, alfalfa is also planted twice on each field so it should have the same total area - 200,000 m2 for each hay and silage.

allisterakun · 2026-07-21T08:05:34Z

As far as I can tell, alfalfa is also planted twice on each field so it should have the same total area - 200,000 m2 for each hay and silage.

Thanks Braden, I have just updated the PR to fix this. Now the total area for both alfalfa (hay and silage) should be 200,000 m2.

What went wrong:

The planned harvest_kill date was after the simulation end date
I divided the field size by the original planned growing period (including extra days after the end of simulation)
Then I clipped the data to only account for days within the simulation window
Result: the daily average was too small → total sum ~172,000

Fix:

Now I divide by the actual growing period within the simulation window, so the daily average and total both match 200,000.

bradenlimb

Looks like it is working correctly to me! Thanks for fixing this!

ew3361zh

Need to spend some more time with the larger functions but thought this feedback would be a good start. Nice work, just multiple complex mappings and loops and exceptions all happening and want to make sure we're defensively guarding properly.

ew3361zh · 2026-07-23T20:02:17Z

                    )
        return values

+    def _scenario_names(self) -> list[str]:


Suggested change

def _scenario_names(self) -> list[str]:

def _get_scenario_names(self) -> list[str]:

or _fetch_scenario_names()

ew3361zh · 2026-07-23T20:06:15Z

+        "cereal_rye_hay": "commodity_prices_wheat_seed_dollar_per_square_meter",
+    }
+
+    _HA_TO_M2: float = 10_000.0


I believe this already exists in general_constants.py and if not, it should be added there.

ew3361zh · 2026-07-23T20:07:27Z

+    _HA_TO_M2: float = 10_000.0
+
+    # Harvest operations that terminate a crop's life in the field.
+    _FINAL_HARVEST_OPS: frozenset[str] = frozenset({"harvest_kill", "kill_only"})


This already exists in harvest_operations.py

ew3361zh · 2026-07-23T20:08:33Z

+        """Return (planting_date, kill_date) pairs for one crop schedule entry.
+
+        Pairs each planting event with its corresponding final harvest
+        (``harvest_kill`` or ``kill_only``).  Intermediate ``harvest_only``
+        operations are ignored.  Pattern expansion (``pattern_repeat``,
+        ``planting_skip``, ``harvesting_skip``) is applied before pairing.


I think a clearer docstring function overview here would be helpful. Also update to numpy style with Returns section, Parameters, etc

ew3361zh · 2026-07-23T20:09:21Z

+    # Harvest operations that terminate a crop's life in the field.
+    _FINAL_HARVEST_OPS: frozenset[str] = frozenset({"harvest_kill", "kill_only"})
+
+    def _growing_periods(


Add action verb to function name. gather e.g.

ew3361zh · 2026-07-23T20:10:29Z

+        raw_p_years: list[int] = list(schedule.get("planting_years") or [])
+        raw_p_days: list[int] = list(schedule.get("planting_days") or [])
+        raw_h_years: list[int] = list(schedule.get("harvest_years") or [])
+        raw_h_days: list[int] = list(schedule.get("harvest_days") or [])
+        raw_h_ops: list[str] = list(schedule.get("harvest_operations") or [])


I would use full words here for the p and h.

ew3361zh · 2026-07-23T20:10:43Z

+        p_years = Schedule.repeat_pattern(raw_p_years, planting_skip, pattern_repeat)
+        p_days = Utility.elongate_list(raw_p_days * (pattern_repeat + 1), len(p_years))
+        h_years = Schedule.repeat_pattern(raw_h_years, harvesting_skip, pattern_repeat)
+        h_days = Utility.elongate_list(raw_h_days * (pattern_repeat + 1), len(h_years))
+        h_ops = Utility.elongate_list(raw_h_ops * (pattern_repeat + 1), len(h_years))


same comment as above.

ew3361zh · 2026-07-23T20:12:12Z

+        For each field, each crop's planting-to-kill periods are located within
+        the simulation window.  The field's area in m² is spread evenly over
+        each growing period (``field_size_m² / period_duration``), then
+        accumulated into a per-simulation-day array keyed by seed commodity.


Add to a Notes section.

ew3361zh · 2026-07-23T20:13:21Z

+        return periods
+
+    def _preprocess_seed_costs(self) -> dict[str, list[float]]:
+        """Build a daily time-series of responsible field area (m²) per seed key.


What is responsible field area?

ew3361zh · 2026-07-23T20:18:12Z

+                plant_date = None
+        return periods
+
+    def _preprocess_seed_costs(self) -> dict[str, list[float]]:


This function is a bit large and doing multiple things - what do you think about extracting some small helper functions?

allisterakun added 4 commits June 24, 2026 13:07

wip

acb3ca1

update

3900c37

Update preprocessing.py

16ab32f

bug fixes

152a503

allisterakun requested a review from bradenlimb June 26, 2026 20:02

update reporting per Braden's feedback

9464d9a

allisterakun linked an issue Jul 2, 2026 that may be closed by this pull request

[Econ] Process seed costs #3080

Open

allisterakun and others added 4 commits July 13, 2026 00:10

cleanup

fd226ed

update

f182bd3

fallback

7898cb2

Merge branch 'economics' into econ_seed_cost_preprocess

ac9c7ae

allisterakun marked this pull request as ready for review July 13, 2026 16:09

fix daily area when clipped

1f3ba02

bradenlimb approved these changes Jul 21, 2026

View reviewed changes

allisterakun requested review from ew3361zh and matthew7838 July 22, 2026 13:52

ew3361zh reviewed Jul 23, 2026

View reviewed changes

	def _scenario_names(self) -> list[str]:
	def _get_scenario_names(self) -> list[str]:

Uh oh!

Conversation

allisterakun commented Jun 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Context

What

Why

How

Test plan

Input Changes

Output Changes

Filter

Uh oh!

bradenlimb commented Jul 20, 2026

Uh oh!

allisterakun commented Jul 20, 2026

Uh oh!

bradenlimb commented Jul 20, 2026

Uh oh!

allisterakun commented Jul 21, 2026

Uh oh!

bradenlimb left a comment

Choose a reason for hiding this comment

Uh oh!

ew3361zh left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

allisterakun commented Jun 24, 2026 •

edited

Loading