Icon

ProvlepsiAstohias(kai)

Predictive Maintenance Workflow

Dataset: ai4i2020.csv

Component 1: Data Import & EDA
Component 2: Feature Engineering
Component 3: Binary Classification (Machine failure)
Component 4: Multiclass Classification (Failure_Type)
Component 5: Hyperparameter Optimization
Component 6: Evaluation & Reporting

EDA

Statistics, Data Explorer, and class distributions for Machine failure and derived Failure_Type.

Feature Engineering

Create Failure_Type, remove identifier columns, compare Dataset A (with Type) vs Dataset B (without Type).

Binary Classification

Decision Tree, Random Forest, Gradient Boosted Trees, k-NN, and evaluation for Machine failure.

Multiclass Classification

Models and evaluation for derived target Failure_Type.

Hyperparameter Optimization

Added a reusable optimization area for classifier tuning. Use parameter loops to maximize F1-score and secondarily inspect Recall. Recommended first pass: optimize tree-based models and k-NN, then extend to Neural Network.

Hyperparameter Optimization Branches

Optimization skeletons for binary classification models: Random Forest, Gradient Boosted Trees, and k-NN. These branches are added first and will be configured in a targeted second pass.

Hyperparameter optimization loop start for Decision Tree on binary training data
Parameter Optimization Loop Start (Table)
Evaluate KNN predictions
Scorer
Decision Tree learner driven by optimization parameters for binary classification
Decision Tree Learner
Score Decision Tree predictions inside binary optimization loop
Scorer
Decision Tree predictor inside optimization loop on binary test data
Decision Tree Predictor
Collect best Decision Tree binary hyperparameters by extracted objective metric
Parameter Optimization Loop End
Extract optimization objective metric from binary scoring table
Cell Extractor
KNN classifier branch for optimization on normalized binary dataset A
K Nearest Neighbor
Random Forest optimization loop start for binary classification
Parameter Optimization Loop Start
Gradient Boosted Trees learner driven by optimization parameters for binary classification
Gradient Boosted Trees Learner
Score KNN predictions inside optimization branch
Scorer
Gradient Boosted Trees predictor inside optimization loop for binary test data
Gradient Boosted Trees Predictor
Score Random Forest predictions inside optimization loop
Scorer
Extract objective metric from KNN scoring table
Cell Extractor
Score Gradient Boosted Trees predictions inside optimization loop
Scorer
Extract objective metric from Gradient Boosted Trees scoring table
Cell Extractor
Extract objective metric from Random Forest scoring table
Cell Extractor
Collect best Random Forest binary hyperparameters
Parameter Optimization Loop End
Collect best KNN binary hyperparameters
Parameter Optimization Loop End
CSV Reader
Collect best Gradient Boosted Trees binary hyperparameters
Parameter Optimization Loop End
Column Filter
Number to String
One hot encode Type for dataset A
One to Many
Decision Tree predictions
Decision Tree Predictor
Table Creator
Normalize dataset A for distance based models
Normalizer
Scorer
Create dataset B without Type
Column Filter
Decision Tree learner for classification
Decision Tree Learner
Train test split dataset A unnormalized
Table Partitioner
Normalize dataset B for distance based models
Normalizer
Train test split dataset B unnormalized
Table Partitioner
Train test split dataset A normalized
Table Partitioner
Decision Tree learner for classification
Decision Tree Learner
Train test split dataset B normalized
Table Partitioner
Decision Tree predictions
Decision Tree Predictor
Interactive data exploration
Data Explorer
Class distribution for Machine failure
Value Counter
EDA statistics for ai4i2020 dataset
Statistics
Create Failure_Type target column from failure indicators
Expression
Remove identifier columns UDI and Product ID
Column Filter
Create Failure_Type target and later encode Type
One to Many
Class distribution for Failure_Type
Value Counter
Random Forest learner for classification
Random Forest Learner
Evaluate Decision Tree predictions
Scorer
Class distribution for derived Failure_Type
Value Counter
Evaluate Random Forest predictions
Scorer
Random Forest predictions
Random Forest Predictor
Gradient Boosted Trees predictions
Gradient Boosted Trees Predictor
Gradient Boosted Trees learner for classification
Gradient Boosted Trees Learner
KNN classifier on normalized dataset A
K Nearest Neighbor
Evaluate Gradient Boosted Trees predictions
Scorer

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: xande
Created at: 2026-05-27
On NodePit since: 2026-06-05
Last update: 2026-06-06
Created with KNIME version: v5.11.0

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!