SDG 6 - Use Case - Improving Water Quality in a Real Home

Use Case: Improving Water Quality in a Real Home

Scenario:

A family of four notices their tap water tastes strange and leaves white stains on utensils. They decide to assess and improve water quality to ensure it meets safe drinking water standards.

---

Step 1: Collect Water Quality Data

A water sample is tested, and the following parameters are recorded:

Parameter	Measured Value	Permissible Limit	Observation
Temperature	30°C	25°C (ideal)	Slightly high, indicating heat exposure.
Turbidity	12 NTU	1-5 NTU	Cloudy water; indicates suspended solids.
TDS	800 mg/L	<500 mg/L (acceptable)	Excess dissolved solids.
pH	6.2	6.5-8.5	Slightly acidic.
Hardness	250 mg/L	<200 mg/L	Hard water; leads to scaling in pipes.
Alkalinity	350 mg/L	200-600 mg/L	Acceptable but needs monitoring.
Nitrates	60 mg/L	<50 mg/L	High; could be due to agricultural runoff.
Chlorides	180 mg/L	<250 mg/L	Within permissible limits.
E. coli Presence	Detected	Absent	Unsafe for drinking without treatment.

---

Step 2: Analyze Data

1. High Turbidity: Indicates suspended particles; could clog filters and reduce water quality.
2. TDS Exceeds Limit: May result in taste issues and health concerns.
3. Low pH: Slightly acidic water may corrode pipes and appliances.
4. Presence of E. coli: Biological contamination makes water unsafe for drinking.
5. High Nitrates: Indicates agricultural contamination, posing health risks (especially for infants).

---

Step 3: Develop a Solution

1. Immediate Steps:

• Install a Water Purification System:

  • RO System: Reduces TDS, nitrates, and hardness.
  • UV Sterilizer: Eliminates E. coli and other pathogens.
  • Activated Carbon Filter: Improves taste and removes organic impurities.

• Boil Water:

  • Boil tap water to kill bacteria until purification systems are installed.

• Prevent Further Contamination:

  • Identify and fix potential contamination sources (e.g., leaking septic tanks).

2. Medium- to Long-Term Measures:

• Rainwater Harvesting:
  • Collect and use rainwater for non-drinking purposes.
• Greywater Recycling:
  • Reuse treated greywater for gardening or cleaning.
• Community Awareness:
  • Educate neighbors about protecting water sources from contamination.

---

Step 4: Re-Test Water Quality

After implementing solutions, retest the water. Sample results might show:

Parameter	Measured Value (After Treatment)	Permissible Limit	Observation
Turbidity	2 NTU	1-5 NTU	Improved; now clear.
TDS	450 mg/L	<500 mg/L	Reduced to acceptable range.
pH	7.2	6.5-8.5	Neutralized; no corrosion risk.
Hardness	120 mg/L	<200 mg/L	No scaling issues.
Nitrates	30 mg/L	<50 mg/L	Reduced; safe for consumption.
E. coli Presence	Not Detected	Absent	Safe for drinking.

---

Step 5: Use Predictive Analytics for Monitoring

Goal: Predict future issues based on current trends.

1. Data Collection: Regularly measure parameters (weekly or monthly) using IoT water quality sensors.

2. Model Prediction: Use machine learning models to predict contamination events (e.g., high TDS or E. coli) based on historical data.

3. Action Plan:

   • Send alerts when contamination is likely.
   • Schedule proactive maintenance for filters.

---

Python Code for Predictive Analytics

Here’s a sample code snippet using Random Forest Regressor to predict TDS based on input parameters.

import pandas as pd
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error

# Sample Data
data = {
    "Turbidity": [12, 8, 5, 3],
    "pH": [6.2, 6.8, 7.0, 7.5],
    "Hardness": [250, 200, 150, 100],
    "Nitrates": [60, 50, 40, 30],
    "TDS": [800, 600, 500, 400],
}
df = pd.DataFrame(data)

# Train-Test Split
X = df.drop("TDS", axis=1)
y = df["TDS"]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)

# Model Training
model = RandomForestRegressor(random_state=42)
model.fit(X_train, y_train)

# Prediction
y_pred = model.predict(X_test)
print("Mean Squared Error:", mean_squared_error(y_test, y_pred))

# Predict for new data
new_sample = pd.DataFrame({"Turbidity": [10], "pH": [7.0], "Hardness": [200], "Nitrates": [50]})
predicted_tds = model.predict(new_sample)
print("Predicted TDS:", predicted_tds[0])

---

Step 6: Mobile App Suggestions

1. Aquagenius:
   • Tracks water usage and quality parameters.
2. Plutonic IoT:
   • Monitors real-time water quality via IoT sensors.