Figure Ilustration AI

FORMOSA NEWS - Tangerang - AI-Powered Decision System Ensures Fair Social Assistance Distribution in Tangerang. A digital breakthrough from Indonesian researchers promises to eliminate favoritism and human bias from the distribution of government welfare programs. Six computer scientists from Universitas Pembangunan Nasional Veteran Jakarta have developed a data-driven Decision Support System designed to make the selection of social assistance recipients fairer, more objective, and highly efficient. The pioneering study, published in the Formosa Journal of Computer and Information Science (FJCIS) in 2026, was conducted by Tengku Rafi Syahrial, Meyrson Agintha Sitepu, Satrio Santoso, Ali Mustofa Izzulhaq, Andreya Naufal Subagyo, and Ati Zaidiah. By testing their algorithmic model in Ciledug, Tangerang City, the Universitas Pembangunan Nasional Veteran Jakarta research team demonstrated that migrating from manual community logging to automated mathematical sorting significantly enhances the accuracy of welfare targeting, ensuring that crucial aid reaches the families who need it most.

Overcoming Bias in Manual Welfare Allocation
Social assistance programs serve as an economic lifeline for underprivileged families, acting as a buffer against financial disparity and unexpected market shocks. However, local governments and community leaders across developing regions often struggle with the logistics of distribution. Traditional screening processes rely heavily on manual data collection and subjective evaluations by neighborhood administrators. This reliance on human discretion creates severe vulnerabilities. Manual systems are frequently plagued by personal biases, data entry inaccuracies, and substantial delays due to limited administrative resources. When aid is distributed inaccurately or perceived as unfair, it triggers widespread community dissatisfaction and leads to a highly inefficient drain on government budgets. To solve this, the Universitas Pembangunan Nasional Veteran Jakarta researchers introduced an objective digital filter that processes community data through a standardized mathematical matrix, neutralizing personal favoritism.

The Mechanics of Simple Additive Weighting
To build a transparent selection framework, the Universitas Pembangunan Nasional Veteran Jakarta team utilized the Simple Additive Weighting (SAW) method. The SAW method is a multi-criteria decision-making algorithm celebrated for its computational simplicity and logical clarity, making it an ideal choice for transparent public governance. Working alongside local administrators in the RT 02, RW 06 neighborhood of Ciledug, Tangerang, the researchers established five core socioeconomic criteria to assess candidate households. Each criterion was assigned a specific mathematical weight reflecting its urgency:
  • Family Income Level (35% weight): Deemed the most critical factor to evaluate immediate economic hardship.
  • Number of Dependents (30% weight): Gauges the financial strain based on the size of the household.
  • Physical Home Conditions (20% weight): Measures structural safety and basic living standards.
  • Employment Status (10% weight): Assesses the long-term income stability of the head of the family.
  • Age of Head of Family (5% weight): Factors in vulnerable demographic groups, such as elderly citizens.
The system functions by converting real-world survey responses into numerical scores. These scores undergo a mathematical process called normalization, which scales all values uniformly. Finally, the normalized values are multiplied by their pre-assigned weights and aggregated into a definitive preference score between 0 and 1 for each household.

Key Findings and Priority Rankings
The algorithmic system was put to the test using actual field data collected from 19 heads of households residing in the Ciledug neighborhood. The algorithm successfully eliminated ambiguity, generating an undisputed priority ranking from the most vulnerable to the least needy candidate.
  • Top Priority Recipient: A resident named Fajar Aji (designated as Alternative A14) achieved the highest preference score of 0.7575. The system flagged his household as the most urgent due to a combination of minimal income and severe structural deficiencies in his residence.
  • Secondary Priorities: Rafi Adrian (A15) followed closely in second place with a score of 0.6466, and Aril Fikrie (A17) secured the third position with a score of 0.6400.
  • Lowest Priority: Conversely, a resident named Adriansyah Sulaiman (A11) placed last in the rankings with a score of 0.3441, as his stable employment and higher relative income indicated lesser dependence on immediate welfare.
The ranking data verified that the Simple Additive Weighting model aligns perfectly with humanitarian goals, consistently prioritizing individuals suffering from severe, compound socioeconomic vulnerabilities.

Real-World Impact and Policy Implications

The implementation of this Decision Support System offers immediate benefits to public policy, municipal governance, and community relations. By deploying an automated framework, local governments can radically accelerate the time required to audit community needs. Furthermore, because the mathematical weights and criteria are publicly verifiable, the system provides an unassailable audit trail that defends local officials against accusations of corruption or nepotism. The researchers from Universitas Pembangunan Nasional Veteran Jakarta emphasize that this framework is highly scalable. It can be integrated directly into broader municipal databases or synchronized with national identity registries to automate welfare updates in real time.

Author Profiles
Tengku Rafi Syahrial is a computer science researcher at Universitas Pembangunan Nasional Veteran Jakarta, specializing in data analytics, decision support algorithms, and digital governance tools.
Meyrson Agintha Sitepu, Satrio Santoso, Ali Mustofa Izzulhaq, Andreya Naufal Subagyo, and Ati Zaidiah are computational research specialists affiliated with Universitas Pembangunan Nasional Veteran Jakarta, focusing on re-engineering software systems and information technology frameworks for social development.

Source
Tengku Rafi Syahrial, Meyrson Agintha Sitepu, Satrio Santoso, Ali Mustofa Izzulhaq, Andreya Naufal Subagyo, Ati Zaidiah. Implementation of Decision Support System with Simple Additive Weighting (SAW) Method for Determination of Social Assistance Recipients: A Case Study in Ciledug, Tangerang. Formosa Journal of Computer and Information Science (FJCIS). Vol 5. No.1 2026. Halaman 157-176.
DOI: https://doi.org/10.55927/fjcis.v5i1.16608
URL: https://journal.formosapublisher.org/index.php/fjcis