Back to Tools
GrowthCommercial

Precision Fish Detection (PFD)

TPS - Total Production Solutions

PFD is a cage-level production-intelligence and biological decision-support platform. It combines stocking, mortality, sampling, growth, feed, FCR, temperature and harvest information to reconstruct production performance, compare actual results against biological expectations and identify inconsistencies that may indicate incorrect stock numbers. Its flagship Missing Fish Detection capability is complemented by growth analysis, model comparison, feeding analysis, KPI monitoring, forecasting and management reporting.

Visit Website

Editorial profile only

User reviews pending

System-generated overview. Cannot be edited by suppliers.

Tool Profile

Profile maintained by the editorial team, not the supplier.Source: precisionfishdetection.comLast reviewed 25 Sept 2026How we rank →

Featured capability

Missing Fish Detection

Detect and quantify hidden stock discrepancies before harvest exposes them.

PFD's flagship capability. It evaluates whether production performance is biologically consistent with the declared fish number and, where evidence supports it, estimates a more plausible population range. It considers starting stock, recorded mortality, sample weights, growth trajectory, feed delivered, model feed requirement, biological FCR, reported FCR, model FCR, reconstructed biomass, harvest information, stock adjustments and observation-to-observation consistency. Rather than treating one abnormal KPI as proof of missing fish, PFD cross-checks multiple biological signals and assesses the strength and consistency of the evidence. The objective is to identify cages where the recorded population has become difficult to reconcile with observed growth and feed performance. Where a discrepancy is supported, PFD estimates a missing-fish range, evaluates confidence and helps identify the period in the production cycle when the discrepancy may have developed. Why this matters: hidden losses may remain undetected until harvest. During that period the farm can be feeding against an inflated fish number, forecasting biomass incorrectly, interpreting FCR using the wrong population and planning harvest tonnage against stock that may no longer be present. Potential explanations can include inaccurate initial stocking, unrecorded mortality, escapes, storm damage, predation, handling losses, transfer discrepancies or harvest-recording errors. PFD uses the available production evidence to help prioritise plausible explanations rather than automatically assigning a single cause. PFD does not replace physical biomass verification or farm-management software. It acts as a biological intelligence and decision-support layer that helps operators identify where verification and management attention should be focused.

Screenshots

Key Capabilities

Missing Fish Intelligence

Find the fish that may exist in the records but no longer exist in the cage.

PFD's flagship analysis tests whether the recorded fish population remains biologically consistent with feed delivered, observed growth, mortality, sampling history, FCR and reconstructed biomass. Where the evidence becomes inconsistent with the recorded population, PFD estimates a plausible missing-fish range and assigns confidence to the finding.

Why it matters

Potential stock discrepancies can be investigated during the production cycle instead of being discovered only when harvest tonnage fails to match expectations.

Loss Window Estimation

Don't just ask how many fish may be missing. Ask when the discrepancy appeared.

PFD analyses the production timeline to identify periods where growth, feed utilisation and recorded population begin to diverge from biological expectations.

Why it matters

Helps management investigate specific operational periods and possible causes such as escapes, storms, mortality, transfers, handling events or recording errors.

Biological Biomass Reconstruction

Reconstruct what the cage biology says should be there.

PFD reconstructs biomass using stocking history, mortality, samples, feed, growth performance and biological models rather than relying only on the latest recorded fish number multiplied by average weight.

Why it matters

Provides an additional biological challenge to the biomass assumptions used for feeding, forecasting and harvest planning.

Growth vs Model

See whether the cage is behaving as expected.

Actual sample weights and growth trajectories are compared against biological growth expectations, making abnormal acceleration, underperformance and unexplained deviations easier to identify.

Why it matters

Turns scattered sample data into a clear biological performance story.

Feed & FCR Intelligence

A good-looking FCR can still be telling you something is wrong.

PFD cross-checks feed delivered, recorded biomass growth, reported FCR, biological FCR and model expectations. Performance that appears unusually good or biologically inconsistent can become evidence that the recorded stock number deserves investigation.

Why it matters

Helps prevent apparently strong KPIs from masking incorrect biomass or population assumptions.

Growth Prognosis

Turn today's cage performance into a forward production view.

PFD projects biological growth using current cage performance and species-specific growth assumptions, allowing operators to examine likely future weight and biomass trajectories.

Why it matters

Supports production planning and provides an independent biological reference for future biomass expectations.

Cage-Level KPI Intelligence

The numbers that matter, without digging through spreadsheets.

PFD consolidates key cage indicators including average weight, recorded fish number, estimated missing fish, feed performance, reported FCR, model FCR, sampling history and data recency.

Why it matters

Allows production managers to identify cages requiring attention quickly rather than manually reconstructing performance from multiple datasets.

Confidence-Based Decision Support

Not every anomaly deserves action.

PFD distinguishes between weak signals, situations requiring additional validation and evidence strong enough to justify operational investigation.

Why it matters

Reduces the risk of reacting to normal biological variation while highlighting cages where multiple independent signals point toward the same problem.

Multi-Cage & Cycle Comparison

One cage can look normal until you compare it with the rest of the farm.

Compare biological performance across cages, production cycles, feeds, origins or management strategies to identify meaningful differences in growth, FCR and stock performance.

Why it matters

Makes farm-wide production data more useful for operational learning and benchmarking.

Existing Data — No New Cage Hardware

Start with the data the farm already produces.

PFD is designed to analyse existing production records including stocking, mortality, feed, samples, temperature, transfers and harvest information.

Why it matters

A farm can investigate a cage without first committing to cameras, sonar systems or additional cage hardware.

What It Helps You Decide

  • Validate whether recorded fish numbers remain biologically plausible
  • Identify cages requiring physical stock verification
  • Detect potential hidden or unrecorded stock losses
  • Assess whether growth is consistent with feed delivered
  • Compare reported FCR with biologically reconstructed FCR
  • Identify periods when stock discrepancies may have developed
  • Evaluate cage-level biological performance
  • Compare actual growth against biological growth models
  • Support feed and biomass planning
  • Improve harvest biomass forecasting
  • Prioritise cages requiring management investigation
  • Compare performance between cages and production cycles

What You Get

Estimated missing-fish range
Stock-discrepancy alerts
Biomass reconstruction
Recorded versus biologically plausible fish-number comparison
Actual versus model growth analysis
Reported versus biological FCR comparison
Feed requirement analysis
Growth prognosis
Potential loss-window estimation
Confidence assessment
Cage-level KPI analysis
Production-cycle comparison
Management investigation guidance
CSV exports
Summary reports

Does not replace

This tool is a decision layer on top of existing operational practice. It is not a substitute for:

  • ·Independent biomass validation
  • ·Health diagnostics
  • ·Oxygen / environmental management

Applicability

Species

sea breamsea basstilapia

Systems

marine cages

Life Stages

juvenileongrowingharvest

Inputs Required

Core Data

  • •Initial stocked fish number
  • •Initial or stocking weight
  • •Mortality records
  • •Feed delivered
  • •Sample dates
  • •Sample average weights

Useful Additional Data

  • •Water temperature
  • •Harvest records
  • •Transfer records
  • •Stock adjustments
  • •Production-cycle dates
  • •Farm growth or feeding models where available

Supported Data Sources

  • •Farm-management system exports
  • •Production spreadsheets
  • •FishTalk data
  • •AquaManager data

Reality Check

Works Well When

  • The farm has reasonably consistent feed, mortality and sampling records
  • Fish are sampled periodically during the production cycle
  • Management suspects biomass or stock-number discrepancies
  • Reported FCR or growth appears inconsistent with field observations
  • The farm wants to investigate discrepancies without installing additional hardware
  • Multiple cages or production cycles are available for comparison

Performs Poorly When

  • Production records contain large gaps or unreliable dates
  • Feed allocation between cages is poorly recorded
  • Sampling is very infrequent or sample weights are unreliable
  • Stock transfers or harvests are not correctly recorded
  • The production history is too short to establish a meaningful biological trajectory
  • The user expects software alone to prove the physical number of fish in a cage

Fit Scores

Species Fit
4
System Fit
4
Data Readiness
3
Complexity
4
Value vs Effort
5

Operational validation

Considering this technology?

Public information can explain what a technology is designed to do, but cannot determine whether it will work effectively on your farm. Validation considers species and system compatibility, scale, existing infrastructure, data availability, integrations, environmental constraints, staff requirements, implementation burden, evidence quality, supplier claims, pilot design and economic exposure.

One question worth testing

“Are there really as many fish in that cage as your records say?”

Most farm-management systems calculate performance from the fish number entered into the system. If that number is wrong, biomass, FCR, feed planning and harvest forecasts can all be calculated from the same incorrect assumption.

PFD independently challenges the recorded population using the biological evidence already contained in the farm's production history.

If the recorded fish number is correct, PFD provides additional confidence in the production record. If it is not, identifying the discrepancy before harvest gives the farm an opportunity to investigate and adjust its decisions earlier.

Entry offer

Give PFD one cage. PFD will analyse whether the recorded stock number is biologically consistent with the cage's feed, growth, mortality and sampling history.

No hardware installation. No replacement of the existing farm-management system. Start with one cage and evaluate the evidence.

Choose a cage you know well — especially one where biomass, FCR, growth or harvest expectations have raised questions — and put the recorded fish number to the test.