391. Are Your Sensors Lying? Fraud Detection in Smart Home Systems Using SAS

Are Your Sensors Lying? Fraud Detection in Smart Home Systems Using SAS

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

HERE IN THIS PROJECT WE USED THESE SAS STATEMENTS —
DATA STEP | PROC SQL |  PROC PRINT | PROC SGPLOT | MACROS | PROC CORR | PROC MEANS | PROC FREQ | PROC UNIVARIATE | APPEND | SET | PROC SORT | MERGE | PROC DATASETS DELETE | DATA FUNCTIONS

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

INTRODUCTION

In today’s digital era, Smart Home Systems have become an integral part of modern living. Smart homes rely on a network of interconnected sensors to monitor, control, and automate various household activities such as lighting, temperature, security, energy consumption, and safety. These sensors continuously generate large volumes of data, which must be analyzed effectively to ensure system reliability, efficiency, and security.

This project, Smart Home Sensors Analytics using SAS, is designed to simulate a real-world IoT (Internet of Things) environment where multiple smart sensors are installed across different rooms of a house. Each sensor records operational metrics such as power consumption, data transmission frequency, error rate, latency, and reliability score. By analyzing this data using SAS, we aim to derive meaningful insights that can help businesses optimize system performance, detect faulty or fraudulent sensors, and plan maintenance activities.

The primary objective of this project is to demonstrate how SAS can be used as a powerful analytics platform for IoT data. The project covers the complete data lifecycle including data creation, data transformation, statistical analysis, visualization, automation using macros, and business interpretation. This makes it a comprehensive case study for students, data analysts, and SAS programmers.

TABLE OF CONTENTS

1. Introduction
2. Business Context
3. Problem Statement
4. Project Objectives
5. Data Description
6. Data Creation using SAS
7. Data Transformation and Cleaning
8. Date and Time Analysis
9. Utilization Classification using Macros
10. Fraud Detection Logic
11. SQL-Based Aggregation
12. Descriptive Statistical Analysis
13. Univariate Analysis
14. Frequency Analysis
15. Correlation Analysis
16. Graphical Visualization
17. Data Transposition
18. Dataset Merging and Appending
19. Character and Numeric Functions
20. Dataset Maintenance and Cleanup
21. Key Business Insights
22. Practical Applications
23. Interview Relevance
24. Conclusion

1. BUSINESS CONTEXT

Smart home technology is widely used by:

• Real estate companies
• IoT device manufacturers
• Home automation service providers
• Energy management firms
• Security and surveillance companies

These organizations need continuous monitoring of sensor behavior to ensure:

• High reliability
• Low power consumption
• Minimal errors
• Fast response time
• Strong security

Poor sensor performance can lead to:

• Increased energy costs
• False alarms
• System downtime
• Customer dissatisfaction
• Security risks

PROBLEM STATEMENT

The key challenges addressed in this project are:

1. Identifying sensors with high power consumption
2. Detecting sensors with high error rates
3. Monitoring latency issues
4. Measuring overall reliability
5. Classifying sensors based on utilization
6. Detecting suspicious or fraudulent sensor behavior
7. Performing statistical and correlation analysis
8. Visualizing performance patterns
9. Managing and cleaning raw sensor data

PROJECT OBJECTIVES

The main objectives of this Smart Home Sensor project are:

• To create a realistic sensor dataset
• To apply SAS data step programming
• To use SQL for business aggregation
• To perform descriptive and inferential statistics
• To implement macros for automation
• To apply date functions for lifecycle analysis
• To visualize data using graphical procedures
• To demonstrate data cleaning techniques
• To build a complete end-to-end analytics pipeline

DATA DESCRIPTION

The dataset consists of the following key variables:

Variable Name	Description
Sensor_ID	Unique identifier for each sensor
Room_Type	Location of sensor
Data_Frequency	Number of data transmissions per minute
Power_Consumption	Energy usage
Error_Rate	Percentage of faulty readings
Latency	Delay in milliseconds
Reliability_Score	Performance rating
Install_Date	Date of installation
Utilization	High/Medium/Low classification
Fraud_Flag	Indicates suspicious behavior
Days_Active	Number of days since installation
Next_Service	Next maintenance date

2. DATASET CREATION

data smart_sensors;

    input Sensor_ID $ Room_Type:$12. Data_Frequency Power_Consumption Error_Rate Latency

          Reliability_Score Install_Date :date9.;

    Power_Pct = Power_Consumption / 100;

    Error_Pct = Error_Rate / 100;

    format Install_Date date9.;

datalines;

S001 LivingRoom 60 35 2 120 95 01JAN2024

S002 Kitchen 30 55 8 200 70 05JAN2024

S003 Bedroom 15 20 1 80 98 10JAN2024

S004 Bathroom 45 40 5 160 85 15JAN2024

S005 Garage 20 70 15 300 60 20JAN2024

S006 Hall 60 25 2 110 96 25JAN2024

S007 Balcony 10 15 1 70 99 01FEB2024

S008 Office 50 60 10 220 75 05FEB2024

S009 Dining 30 45 6 180 88 10FEB2024

S010 KidsRoom 20 30 3 100 94 15FEB2024

S011 StoreRoom 10 10 0.5 50 100 20FEB2024

S012 Kitchen 25 65 12 240 72 25FEB2024

S013 LivingRoom 55 38 4 150 90 01MAR2024

S014 Bedroom 15 22 1 90 97 05MAR2024

S015 Garage 30 75 18 350 55 10MAR2024

S016 Office 60 58 9 210 78 15MAR2024

;

run;

proc print data=smart_sensors;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090

3. MACRO – UTILIZATION CLASSIFICATION

%macro utilization;

data smart_sensors;

    set smart_sensors;

    length Utilization $8.;

    if Power_Consumption > 60 then Utilization = "High";

    else if 30 < Power_Consumption < 60 then Utilization = "Medium";

    else Utilization = "Low";

run;

proc print data=smart_sensors;

run;

%mend;

%utilization;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct	Utilization
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020	Medium
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080	Medium
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010	Low
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050	Medium
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150	High
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020	Low
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010	Low
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100	Low
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060	Medium
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030	Low
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005	Low
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120	High
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040	Medium
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010	Low
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180	High
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090	Medium

4. DATE FUNCTIONS

data smart_sensors;

    set smart_sensors;

    Today = today();

    Days_Active = intck('day', Install_Date, Today);

    Next_Service = intnx('month', Install_Date, 6);

run;

proc print data=smart_sensors;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct	Utilization	Today	Days_Active	Next_Service
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020	Medium	24141	765	23558
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080	Medium	24141	761	23558
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010	Low	24141	756	23558
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050	Medium	24141	751	23558
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150	High	24141	746	23558
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020	Low	24141	741	23558
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010	Low	24141	734	23589
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100	Low	24141	730	23589
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060	Medium	24141	725	23589
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030	Low	24141	720	23589
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005	Low	24141	715	23589
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120	High	24141	710	23589
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040	Medium	24141	705	23620
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010	Low	24141	701	23620
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180	High	24141	696	23620
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090	Medium	24141	691	23620

5. FRAUD DETECTION MACRO

%macro fraud_check;

data fraud_sensors;

    set smart_sensors;

    length Fraud_Flag $3.;

    if Error_Rate > 10 or Latency > 250 then Fraud_Flag = "YES";

    else Fraud_Flag = "NO";

run;

proc print data=fraud_sensors;

run;

%mend;

%fraud_check;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct	Utilization	Today	Days_Active	Next_Service	Fraud_Flag
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020	Medium	24141	765	23558	NO
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080	Medium	24141	761	23558	NO
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010	Low	24141	756	23558	NO
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050	Medium	24141	751	23558	NO
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150	High	24141	746	23558	YES
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020	Low	24141	741	23558	NO
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010	Low	24141	734	23589	NO
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100	Low	24141	730	23589	NO
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060	Medium	24141	725	23589	NO
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030	Low	24141	720	23589	NO
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005	Low	24141	715	23589	NO
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120	High	24141	710	23589	YES
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040	Medium	24141	705	23620	NO
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010	Low	24141	701	23620	NO
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180	High	24141	696	23620	YES
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090	Medium	24141	691	23620	NO

6. PROC SQL ANALYTICS

proc sql;

create table room_summary as

select Room_Type,

       avg(Power_Consumption) as Avg_Power,

       avg(Error_Rate) as Avg_Error,

       avg(Reliability_Score) as Avg_Reliability

from smart_sensors

group by Room_Type;

quit;

proc print data=room_summary;

run;

OUTPUT:

Obs	Room_Type	Avg_Power	Avg_Error	Avg_Reliability
1	Balcony	15.0	1.0	99.0
2	Bathroom	40.0	5.0	85.0
3	Bedroom	21.0	1.0	97.5
4	Dining	45.0	6.0	88.0
5	Garage	72.5	16.5	57.5
6	Hall	25.0	2.0	96.0
7	KidsRoom	30.0	3.0	94.0
8	Kitchen	60.0	10.0	71.0
9	LivingRoom	36.5	3.0	92.5
10	Office	59.0	9.5	76.5
11	StoreRoom	10.0	0.5	100.0

7. PROC MEANS

proc means data=smart_sensors mean min max std;

var Power_Consumption Error_Rate Latency Reliability_Score;

run;

OUTPUT:

The MEANS Procedure

Variable	Mean	Minimum	Maximum	Std Dev
Power_Consumption Error_Rate Latency Reliability_Score	41.4375000 6.0937500 164.3750000 84.5000000	10.0000000 0.5000000 50.0000000 55.0000000	75.0000000 18.0000000 350.0000000 100.0000000	20.4840385 5.4228490 85.3204743 14.4545264

8. PROC UNIVARIATE

proc univariate data=smart_sensors;

    var Power_Consumption Latency;

    histogram Power_Consumption Latency / normal kernel midpoints=10 to 80 by 10 

                                          cfill=gray vscale=count;

    inset mean median std min max n / position=ne;

    probplot Power_Consumption Latency / normal(mu=est sigma=est);

run;

OUTPUT:

The UNIVARIATE Procedure

Variable: Power_Consumption

Moments
N	16	Sum Weights	16
Mean	41.4375	Sum Observations	663
Std Deviation	20.4840385	Variance	419.595833
Skewness	0.130103	Kurtosis	-1.2290794
Uncorrected SS	33767	Corrected SS	6293.9375
Coeff Variation	49.4335771	Std Error Mean	5.12100963

Basic Statistical Measures
Location		Variability
Mean	41.43750	Std Deviation	20.48404
Median	39.00000	Variance	419.59583
Mode	.	Range	65.00000
		Interquartile Range	35.50000

Tests for Location: Mu0=0
Test	Statistic		p Value
Student's t	t	8.091666	Pr > |t|	<.0001
Sign	M	8	Pr >= |M|	<.0001
Signed Rank	S	68	Pr >= |S|	<.0001

Quantiles (Definition 5)
Level	Quantile
100% Max	75.0
99%	75.0
95%	75.0
90%	70.0
75% Q3	59.0
50% Median	39.0
25% Q1	23.5
10%	15.0
5%	10.0
1%	10.0
0% Min	10.0

Extreme Observations
Lowest		Highest
Value	Obs	Value	Obs
10	11	58	16
15	7	60	8
20	3	65	12
22	14	70	5
25	6	75	15

The UNIVARIATE Procedure

The UNIVARIATE Procedure

Fitted Normal Distribution for Power_Consumption

Parameters for Normal Distribution
Parameter	Symbol	Estimate
Mean	Mu	41.4375
Std Dev	Sigma	20.48404

Goodness-of-Fit Tests for Normal Distribution
Test	Statistic		p Value
Kolmogorov-Smirnov	D	0.12104672	Pr > D	>0.150
Cramer-von Mises	W-Sq	0.03880802	Pr > W-Sq	>0.250
Anderson-Darling	A-Sq	0.25231907	Pr > A-Sq	>0.250

Quantiles for Normal Distribution
Percent	Quantile
	Observed	Estimated
1.0	10.00000	-6.21550
5.0	10.00000	7.74425
10.0	15.00000	15.18615
25.0	23.50000	27.62123
50.0	39.00000	41.43750
75.0	59.00000	55.25377
90.0	70.00000	67.68885
95.0	75.00000	75.13075
99.0	75.00000	89.09050

The UNIVARIATE Procedure

The UNIVARIATE Procedure

Variable: Latency

Moments
N	16	Sum Weights	16
Mean	164.375	Sum Observations	2630
Std Deviation	85.3204743	Variance	7279.58333
Skewness	0.70811576	Kurtosis	-0.0698201
Uncorrected SS	541500	Corrected SS	109193.75
Coeff Variation	51.905992	Std Error Mean	21.3301186

Basic Statistical Measures
Location		Variability
Mean	164.3750	Std Deviation	85.32047
Median	155.0000	Variance	7280
Mode	.	Range	300.00000
		Interquartile Range	120.00000

Tests for Location: Mu0=0
Test	Statistic		p Value
Student's t	t	7.706239	Pr > |t|	<.0001
Sign	M	8	Pr >= |M|	<.0001
Signed Rank	S	68	Pr >= |S|	<.0001

Quantiles (Definition 5)
Level	Quantile
100% Max	350
99%	350
95%	350
90%	300
75% Q3	215
50% Median	155
25% Q1	95
10%	70
5%	50
1%	50
0% Min	50

Extreme Observations
Lowest		Highest
Value	Obs	Value	Obs
50	11	210	16
70	7	220	8
80	3	240	12
90	14	300	5
100	10	350	15

The UNIVARIATE Procedure

The UNIVARIATE Procedure

Fitted Normal Distribution for Latency

Parameters for Normal Distribution
Parameter	Symbol	Estimate
Mean	Mu	164.375
Std Dev	Sigma	85.32047

Goodness-of-Fit Tests for Normal Distribution
Test	Statistic		p Value
Kolmogorov-Smirnov	D	0.13600233	Pr > D	>0.150
Cramer-von Mises	W-Sq	0.04068360	Pr > W-Sq	>0.250
Anderson-Darling	A-Sq	0.29784435	Pr > A-Sq	>0.250

Quantiles for Normal Distribution
Percent	Quantile
	Observed	Estimated
1.0	50.0000	-34.1101
5.0	50.0000	24.0353
10.0	70.0000	55.0324
25.0	95.0000	106.8272
50.0	155.0000	164.3750
75.0	215.0000	221.9228
90.0	300.0000	273.7176
95.0	350.0000	304.7147
99.0	350.0000	362.8601

The UNIVARIATE Procedure

9. PROC FREQ

proc freq data=smart_sensors;

tables Room_Type*Utilization;

run;

OUTPUT:

The FREQ Procedure

Frequency Percent Row Pct Col Pct

Table of Room_Type by Utilization Room_Type Utilization High Low Medium Total Balcony 0 0.00 0.00 0.00 1 6.25 100.00 14.29 0 0.00 0.00 0.00 1 6.25     Bathroom 0 0.00 0.00 0.00 0 0.00 0.00 0.00 1 6.25 100.00 16.67 1 6.25     Bedroom 0 0.00 0.00 0.00 2 12.50 100.00 28.57 0 0.00 0.00 0.00 2 12.50     Dining 0 0.00 0.00 0.00 0 0.00 0.00 0.00 1 6.25 100.00 16.67 1 6.25     Garage 2 12.50 100.00 66.67 0 0.00 0.00 0.00 0 0.00 0.00 0.00 2 12.50     Hall 0 0.00 0.00 0.00 1 6.25 100.00 14.29 0 0.00 0.00 0.00 1 6.25     KidsRoom 0 0.00 0.00 0.00 1 6.25 100.00 14.29 0 0.00 0.00 0.00 1 6.25     Kitchen 1 6.25 50.00 33.33 0 0.00 0.00 0.00 1 6.25 50.00 16.67 2 12.50     LivingRoom 0 0.00 0.00 0.00 0 0.00 0.00 0.00 2 12.50 100.00 33.33 2 12.50     Office 0 0.00 0.00 0.00 1 6.25 50.00 14.29 1 6.25 50.00 16.67 2 12.50     StoreRoom 0 0.00 0.00 0.00 1 6.25 100.00 14.29 0 0.00 0.00 0.00 1 6.25     Total 3 18.75 7 43.75 6 37.50 16 100.00

10. PROC CORR

proc corr data=smart_sensors;

var Power_Consumption Error_Rate Latency Reliability_Score;

run;

OUTPUT:

The CORR Procedure

4 Variables:	Power_Consumption Error_Rate Latency Reliability_Score

Simple Statistics
Variable	N	Mean	Std Dev	Sum	Minimum	Maximum
Power_Consumption	16	41.43750	20.48404	663.00000	10.00000	75.00000
Error_Rate	16	6.09375	5.42285	97.50000	0.50000	18.00000
Latency	16	164.37500	85.32047	2630	50.00000	350.00000
Reliability_Score	16	84.50000	14.45453	1352	55.00000	100.00000

Pearson Correlation Coefficients, N = 16 Prob > |r| under H0: Rho=0
	Power_Consumption	Error_Rate	Latency	Reliability_Score
Power_Consumption	1.00000	0.95776 <.0001	0.97382 <.0001	-0.95479 <.0001
Error_Rate	0.95776 <.0001	1.00000	0.98678 <.0001	-0.97957 <.0001
Latency	0.97382 <.0001	0.98678 <.0001	1.00000	-0.97438 <.0001
Reliability_Score	-0.95479 <.0001	-0.97957 <.0001	-0.97438 <.0001	1.00000

11. PROC SGPLOT

proc sgplot data=smart_sensors;

scatter x=Power_Consumption y=Reliability_Score;

run;

OUTPUT:

12. PROC TRANSPOSE

proc transpose data=smart_sensors out=transposed;

by Sensor_ID NotSorted;

var Power_Consumption Error_Rate Latency;

run;

proc print data=transposed;

run;

OUTPUT:

Obs	Sensor_ID	_NAME_	COL1
1	S001	Power_Consumption	35.0
2	S001	Error_Rate	2.0
3	S001	Latency	120.0
4	S002	Power_Consumption	55.0
5	S002	Error_Rate	8.0
6	S002	Latency	200.0
7	S003	Power_Consumption	20.0
8	S003	Error_Rate	1.0
9	S003	Latency	80.0
10	S004	Power_Consumption	40.0
11	S004	Error_Rate	5.0
12	S004	Latency	160.0
13	S005	Power_Consumption	70.0
14	S005	Error_Rate	15.0
15	S005	Latency	300.0
16	S006	Power_Consumption	25.0
17	S006	Error_Rate	2.0
18	S006	Latency	110.0
19	S007	Power_Consumption	15.0
20	S007	Error_Rate	1.0
21	S007	Latency	70.0
22	S008	Power_Consumption	60.0
23	S008	Error_Rate	10.0
24	S008	Latency	220.0
25	S009	Power_Consumption	45.0
26	S009	Error_Rate	6.0
27	S009	Latency	180.0
28	S010	Power_Consumption	30.0
29	S010	Error_Rate	3.0
30	S010	Latency	100.0
31	S011	Power_Consumption	10.0
32	S011	Error_Rate	0.5
33	S011	Latency	50.0
34	S012	Power_Consumption	65.0
35	S012	Error_Rate	12.0
36	S012	Latency	240.0
37	S013	Power_Consumption	38.0
38	S013	Error_Rate	4.0
39	S013	Latency	150.0
40	S014	Power_Consumption	22.0
41	S014	Error_Rate	1.0
42	S014	Latency	90.0
43	S015	Power_Consumption	75.0
44	S015	Error_Rate	18.0
45	S015	Latency	350.0
46	S016	Power_Consumption	58.0
47	S016	Error_Rate	9.0
48	S016	Latency	210.0

13. APPEND, MERGE, SET

data new_sensors;

  input Sensor_ID $ Room_Type $ Data_Frequency Power_Consumption Error_Rate Latency 

      Reliability_Score Install_Date :date9.;

  datalines;

  S017 Hall 20 28 2 95 96 20MAR2024

  S018 Office 50 62 11 230 74 25MAR2024

;

run;

proc print data=new_sensors;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date
1	S017	Hall	20	28	2	95	96	23455
2	S018	Office	50	62	11	230	74	23460

proc append base=smart_sensors 

            data=new_sensors;

run;

proc print data=smart_sensors;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct	Utilization	Today	Days_Active	Next_Service
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020	Medium	24141	765	23558
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080	Medium	24141	761	23558
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010	Low	24141	756	23558
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050	Medium	24141	751	23558
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150	High	24141	746	23558
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020	Low	24141	741	23558
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010	Low	24141	734	23589
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100	Low	24141	730	23589
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060	Medium	24141	725	23589
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030	Low	24141	720	23589
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005	Low	24141	715	23589
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120	High	24141	710	23589
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040	Medium	24141	705	23620
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010	Low	24141	701	23620
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180	High	24141	696	23620
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090	Medium	24141	691	23620
17	S017	Hall	20	28	2.0	95	96	20MAR2024	.	.		.	.	.
18	S018	Office	50	62	11.0	230	74	25MAR2024	.	.		.	.	.

proc sort data=smart_sensors;by Sensor_ID;run;

proc print data=smart_sensors;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct	Utilization	Today	Days_Active	Next_Service
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020	Medium	24141	765	23558
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080	Medium	24141	761	23558
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010	Low	24141	756	23558
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050	Medium	24141	751	23558
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150	High	24141	746	23558
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020	Low	24141	741	23558
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010	Low	24141	734	23589
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100	Low	24141	730	23589
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060	Medium	24141	725	23589
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030	Low	24141	720	23589
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005	Low	24141	715	23589
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120	High	24141	710	23589
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040	Medium	24141	705	23620
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010	Low	24141	701	23620
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180	High	24141	696	23620
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090	Medium	24141	691	23620
17	S017	Hall	20	28	2.0	95	96	20MAR2024	.	.		.	.	.
18	S018	Office	50	62	11.0	230	74	25MAR2024	.	.		.	.	.

proc sort data=fraud_sensors;by Sensor_ID;run;

proc print data=fraud_sensors;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct	Utilization	Today	Days_Active	Next_Service	Fraud_Flag
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020	Medium	24141	765	23558	NO
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080	Medium	24141	761	23558	NO
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010	Low	24141	756	23558	NO
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050	Medium	24141	751	23558	NO
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150	High	24141	746	23558	YES
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020	Low	24141	741	23558	NO
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010	Low	24141	734	23589	NO
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100	Low	24141	730	23589	NO
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060	Medium	24141	725	23589	NO
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030	Low	24141	720	23589	NO
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005	Low	24141	715	23589	NO
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120	High	24141	710	23589	YES
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040	Medium	24141	705	23620	NO
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010	Low	24141	701	23620	NO
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180	High	24141	696	23620	YES
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090	Medium	24141	691	23620	NO

data merged;

merge smart_sensors 

      fraud_sensors;

by Sensor_ID;

run;

proc print data=merged;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Data_Frequency	Power_Consumption	Error_Rate	Latency	Reliability_Score	Install_Date	Power_Pct	Error_Pct	Utilization	Today	Days_Active	Next_Service	Fraud_Flag
1	S001	LivingRoom	60	35	2.0	120	95	01JAN2024	0.35	0.020	Medium	24141	765	23558	NO
2	S002	Kitchen	30	55	8.0	200	70	05JAN2024	0.55	0.080	Medium	24141	761	23558	NO
3	S003	Bedroom	15	20	1.0	80	98	10JAN2024	0.20	0.010	Low	24141	756	23558	NO
4	S004	Bathroom	45	40	5.0	160	85	15JAN2024	0.40	0.050	Medium	24141	751	23558	NO
5	S005	Garage	20	70	15.0	300	60	20JAN2024	0.70	0.150	High	24141	746	23558	YES
6	S006	Hall	60	25	2.0	110	96	25JAN2024	0.25	0.020	Low	24141	741	23558	NO
7	S007	Balcony	10	15	1.0	70	99	01FEB2024	0.15	0.010	Low	24141	734	23589	NO
8	S008	Office	50	60	10.0	220	75	05FEB2024	0.60	0.100	Low	24141	730	23589	NO
9	S009	Dining	30	45	6.0	180	88	10FEB2024	0.45	0.060	Medium	24141	725	23589	NO
10	S010	KidsRoom	20	30	3.0	100	94	15FEB2024	0.30	0.030	Low	24141	720	23589	NO
11	S011	StoreRoom	10	10	0.5	50	100	20FEB2024	0.10	0.005	Low	24141	715	23589	NO
12	S012	Kitchen	25	65	12.0	240	72	25FEB2024	0.65	0.120	High	24141	710	23589	YES
13	S013	LivingRoom	55	38	4.0	150	90	01MAR2024	0.38	0.040	Medium	24141	705	23620	NO
14	S014	Bedroom	15	22	1.0	90	97	05MAR2024	0.22	0.010	Low	24141	701	23620	NO
15	S015	Garage	30	75	18.0	350	55	10MAR2024	0.75	0.180	High	24141	696	23620	YES
16	S016	Office	60	58	9.0	210	78	15MAR2024	0.58	0.090	Medium	24141	691	23620	NO
17	S017	Hall	20	28	2.0	95	96	20MAR2024	.	.		.	.	.
18	S018	Office	50	62	11.0	230	74	25MAR2024	.	.		.	.	.

14. CHARACTER FUNCTIONS

data smart_sensors;

    set smart_sensors;

    Sensor_Clean = strip(Sensor_ID);

    Room_Upper = upcase(Room_Type);

    Room_Lower = lowcase(Room_Type);

    Room_Proper = propcase(Room_Type);

    Label = catx("-", Sensor_ID, Room_Type);

run;

proc print data=smart_sensors;

 var Sensor_ID Room_Type Sensor_Clean Room_Upper Room_Lower Label;

run;

OUTPUT:

Obs	Sensor_ID	Room_Type	Sensor_Clean	Room_Upper	Room_Lower	Label
1	S001	LivingRoom	S001	LIVINGROOM	livingroom	S001-LivingRoom
2	S002	Kitchen	S002	KITCHEN	kitchen	S002-Kitchen
3	S003	Bedroom	S003	BEDROOM	bedroom	S003-Bedroom
4	S004	Bathroom	S004	BATHROOM	bathroom	S004-Bathroom
5	S005	Garage	S005	GARAGE	garage	S005-Garage
6	S006	Hall	S006	HALL	hall	S006-Hall
7	S007	Balcony	S007	BALCONY	balcony	S007-Balcony
8	S008	Office	S008	OFFICE	office	S008-Office
9	S009	Dining	S009	DINING	dining	S009-Dining
10	S010	KidsRoom	S010	KIDSROOM	kidsroom	S010-KidsRoom
11	S011	StoreRoom	S011	STOREROOM	storeroom	S011-StoreRoom
12	S012	Kitchen	S012	KITCHEN	kitchen	S012-Kitchen
13	S013	LivingRoom	S013	LIVINGROOM	livingroom	S013-LivingRoom
14	S014	Bedroom	S014	BEDROOM	bedroom	S014-Bedroom
15	S015	Garage	S015	GARAGE	garage	S015-Garage
16	S016	Office	S016	OFFICE	office	S016-Office
17	S017	Hall	S017	HALL	hall	S017-Hall
18	S018	Office	S018	OFFICE	office	S018-Office

15. PROC DATASETS DELETE

proc datasets library=work;

  delete transposed;

run;

quit;

LOG:

NOTE: Deleting WORK.TRANSPOSED (memtype=DATA).

16. KEY BUSINESS INSIGHTS

From the analysis of smart home sensor data, several important insights can be derived:

• Sensors installed in Garages and Kitchens tend to consume more power.
• Higher latency generally leads to lower reliability scores.
• Sensors with error rates above 10% show suspicious behavior and require investigation.
• Older sensors show declining performance compared to recently installed ones.
• Rooms with more automation devices generate higher data frequency.

These insights help organizations optimize infrastructure and improve customer satisfaction.

17. PRACTICAL APPLICATIONS

This type of analytics system is used in real-world scenarios such as:

• Smart city monitoring
• Energy management systems
• Hospital automation
• Industrial IoT platforms
• Security surveillance systems
• Predictive maintenance systems

18. INTERVIEW AND CAREER RELEVANCE

This project is highly valuable for:

• SAS Programmer interviews
• Data Analyst roles
• IoT analytics positions
• Business intelligence profiles

It demonstrates skills in:

• Data engineering
• Statistical analysis
• SQL programming
• Automation
• Visualization
• Business interpretation

Recruiters prefer candidates who can explain:

• Business problem
• Data logic
• Technical implementation
• Insights and conclusions

This project fulfills all those expectations.

CONCLUSION

The Smart Home Sensors Analytics using SAS project provides a complete demonstration of how real-world IoT data can be analyzed using SAS. Starting from data creation to final business insights, the project covers every important stage of the analytics lifecycle.

By implementing statistical procedures, SQL queries, macros, and visualization techniques, this project successfully transforms raw sensor data into actionable intelligence. It highlights the importance of monitoring sensor performance, detecting anomalies, and ensuring system reliability in smart environments.

From a learning perspective, this project strengthens understanding of core SAS concepts such as data steps, procedures, macros, functions, and reporting. From a business perspective, it shows how data analytics supports operational efficiency, cost reduction, risk management, and strategic planning.

In summary, this project acts as a bridge between technical SAS programming skills and real-world business applications, making it an ideal case study for both academic learning and professional development.

INTERVIEW QUESTIONS FOR YOU

·  What is %DO and %END?

·  What is %PUT?

·  What is the difference between local and global macro variables?

 

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

About the Author:

SAS Learning Hub is a data analytics and SAS programming platform focused on clinical, financial, and real-world data analysis. The content is created by professionals with academic training in Pharmaceutics and hands-on experience in Base SAS, PROC SQL, Macros, SDTM, and ADaM, providing practical and industry-relevant SAS learning resources.

Disclaimer:

The datasets and analysis in this article are created for educational and demonstration purposes only. They do not represent SMART HOME data.

Our Mission:

This blog provides industry-focused SAS programming tutorials and analytics projects covering finance, healthcare, and technology.

This project is suitable for:

·  Students learning SAS

·  Data analysts building portfolios

·  Professionals preparing for SAS interviews

·  Bloggers writing about analytics and smart cities

·  EV and energy industry professionals

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Follow Us On : 

 

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

--->Follow our blog for more SAS-based analytics projects and industry data models.

---> Support Us By Following Our Blog..

To deepen your understanding of SAS analytics, please refer to our other data science and industry-focused projects listed below:

1.Can SAS Predict Airline Disasters Before They Happen?

2.Which Indian Cuisine Dominates the Restaurant Market? Complete Sas Data

3.Which Is the Most Spoken Language in India?Complete Sas Data

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

About Us | Contact | Privacy Policy