BUSINESS INTELLIGENCE

Project Overview
This project aims to forecast passenger numbers and freight volumes at Purwokerto Station using the Exponential Triple Smoothing (ETS) model. The project involves collecting historical data, performing time series analysis, and implementing the ETS method to predict future trends. The predictions will help in optimizing operational strategies and resource allocation at the station. A key component of this project is the creation of an interactive dashboard using Microsoft Excel, which will visualize the forecasting results in real time. 

Methodology
The project will involve two stages of forecasting:

• Manual Calculations: ETS will be manually applied to historical data, with predictions assessed for accuracy using the Mean Absolute Percentage Error (MAPE). We aim for MAPE values under 30%, indicating a high level of prediction accuracy.

• Excel Implementation: The same methodology will be implemented using Microsoft Excel formulas to streamline calculations and improve prediction efficiency. The forecasts will be updated monthly to track trends and seasonality accurately.

Key Insights

• Accurate Forecasting: The ETS model will provide predictions for both passenger and freight data with a target accuracy of 80% based on past performance.

• Interactive Dashboard: An Excel dashboard will visualize predictions, with features such as monthly data updates and trend forecasting for the next 10 years. KPIs will be incorporated, focusing on operational efficiency and accuracy.

• Real-time Data Analysis: The dashboard will allow station managers to adjust forecasting parameters in real-time and monitor key performance indicators, with the goal of improving decision-making speed by 50% and reducing operational inefficiencies by 30%.
  
  

This Dashboard Can:

• Provide accurate forecasts with less than 15% error margin based on MAPE.

• Track and compare forecast data with actual outcomes, visualizing discrepancies under 15%.

• Allow for real-time adjustments and improvements to predictions, with a reduction in forecasting time by 60%.

• Visualize monthly performance indicators such as passenger counts and freight volumes to optimize station operations.

Project Overview
This project analyzes sales, products, and customer data from an e-commerce platform to enhance strategic decision-making. The analysis focused on improving marketing strategies, optimizing inventory, and enhancing customer service. Using Microsoft Excel, the datasets were cleaned, integrated, and visualized to provide actionable insights.

Methodology

1. Data Collection

• Datasets in CSV format (sales, products, and customers) were imported into Microsoft Excel.

• Data transformation was done using Power Query, ensuring consistent column headers and data types.

2. Data Cleaning

• Missing values were handled using statistical methods: median for numeric data (e.g., age) and mode for categorical data (e.g., gender).

• Issues such as duplicate IDs, invalid dates, and extreme outliers in sales quantities were resolved.

3. Data Integration and Visualization

• Datasets were combined into a single table using relational keys (e.g., customer IDs, product IDs).

• Pivot Tables and Charts were utilized to analyze trends, demographics, and product performance.

• Interactive slicers enabled dynamic filtering by variables like time, product categories, and customer demographics.

Key Insights

1. Sales Trends

• Total revenue reached $14.06 billion, with stable monthly sales averaging 238 units.

• Seasonal spikes in sales during holidays presented opportunities for targeted promotions.

2. Customer Demographics

• Most customers were aged 27–36 (25%), followed by 37–46 (18%).

• Gender distribution was nearly equal, suggesting balanced marketing efforts.

3. Product Analysis

• Top products: Sweaters (2,739 units) and scarves (2,659 units).

• Low-performing products like jeans and sunglasses require re-evaluation.

4. Location Analysis

• Major sales hubs were Jakarta, Surabaya, and Medan, highlighting the need for optimized logistics and localized campaigns.

Project Overview:

This project involved creating an interactive dashboard using Google Looker Studio to extract actionable insights from a platform's dataset, including user demographics, survey participation, and payment trends. The dataset consisted of user profiles, surveys, participation records, and payments, which were integrated and analyzed to deliver a comprehensive overview of platform activity.

Key Insights:

• User Demographics: Most users hold Bachelor’s, Master’s, or PhD degrees, with dominant groups including students, teachers, and professionals. These insights inform marketing and user engagement strategies.

• Survey Participation Trends: Participation fluctuated over time, peaking in early 2023, and generally trended downward. Targeted engagement efforts can help sustain or increase participation.

• Payment Trends: Payments mirrored participation trends, peaking in early 2023. The highest payments ranged from $1,300 to $1,850, suggesting areas to optimize compensation strategies.

• Survey Efficiency: Actual participation was mostly aligned with planned targets, but some variation indicates room for improvement in survey targeting.

• Reward vs. Participation: Higher reward amounts correlated with increased participation, highlighting the impact of competitive compensation.
  
  

Methodology:

• Data Integration & Cleaning: Combined user, survey, participation, and payment data, ensuring consistency and proper relationships between datasets.

• Exploratory Data Analysis: Analyzed the data to uncover trends and key metrics related to demographics, participation, and payments.

• Visualization Design: Developed interactive visualizations (e.g., bar charts, pie charts) to highlight key metrics, trends, and comparisons, using filters and drill-downs for stakeholder exploration.

This project presents a thorough data-driven analysis of business performance, leveraging real-world data across multiple dimensions—orders, sales channels, industries, suppliers, and key personnel (PICs). The objective of the study is to uncover actionable insights, identify areas for improvement, and highlight success drivers that contribute to profitability and long-term sustainability.

Through this analysis, high-performing channels such as Cashured and Fashionista were identified for their significant impact on both revenue and profit. Conversely, underperforming channels like Metro Bank and Dinostore were analyzed to understand their inefficiencies, providing targeted recommendations for improvement.

The project also highlights key trends within the Finance and Retail industries, which contribute 88% of total profits, emphasizing the importance of diversification to mitigate risks. Additionally, top-performing personnel such as Ismail Salahudin were recognized for their contributions to profitability, underscoring the value of effective leadership and operational efficiency.

The insights generated from this project serve as a strategic guide for optimizing business operations, reallocating resources, and aligning efforts to maximize profitability. This portfolio exemplifies how a structured approach to data analysis can drive sustainable business growth and ensure resilience in a competitive market landscape.

Project Overview:
Developed a comprehensive HCM Analyst Dashboard, providing real-time insights into human capital metrics such as employee demographics, skills development, and workforce performance. This project aimed to empower HR teams with data-driven decision-making tools for effective talent management.

Key Responsibilities:

• Collected and analyzed data from 49 interns across 12 regions in Indonesia.

• Designed and implemented an interactive dashboard visualizing key metrics and trends.

• Collaborated with cross-functional teams to ensure data accuracy and actionable insights.
  
  

Dashboard Components:

• Intern Demographics: Visualized geographic distribution, analyzed age distribution (average: 21), and gender balance (55.10% Female, 44.90% Male).

• Skills and Divisions: Broke down intern distribution across 5 divisions, tracked skills development trends.

• Social Media Impact: Monitored collective reach (44,994 Instagram followers, 12,590 LinkedIn connections).

• Work Preferences: Visualized work model preferences (65.3% Remote, 30.6% Hybrid, 4.1% On-site).

• Training and Development: Tracked hands-on and online training preferences, and monitored interest in roles such as trainer and MC/moderator.
  
  

Achievements:

• Increased data visibility by 100%, enabling real-time access to program metrics.

• Identified trends that led to a 20% improvement in intern placement efficiency.

• Boosted intern satisfaction rates by 15% through data-driven program enhancements.

In this project, I designed and developed a comprehensive Sales Analytics Dashboard, enabling real-time monitoring of key performance indicators (KPIs) and trends across various dimensions of the business. The dashboard provided actionable insights for decision-making, resulting in improved sales strategies and profit optimization.

Key Features:

• High-Level KPI Overview: Showcased Total Sales of $2.3M, Profit of $286K, and a Profit Margin of 12.47%.

• Monthly Performance Tracker: Revealed peak sales month (November: $352K) and highest profit margin month (February: 17.23%).

• Segment Analysis: Highlighted the Consumer segment as the leading contributor with 50% of total sales.

• Product Performance: Identified the top 10 products by sales, with the Canon imageCLASS 2200 Advanced Copier generating the highest sales ($61.6K).

• Multi-dimensional Filters: Enabled data exploration across Year, Segment, State, Region, and Month through dynamic filtering.
  
  

Measured Achievements:

• Increased data accessibility, reducing report generation time by 75%.

• Identified a 20% profit margin improvement opportunity in underperforming months.

• Boosted sales for top-performing products by 15% through targeted marketing strategies.

• Enabled real-time decision-making, contributing to a 10% increase in annual profits.
  
  

Methodologies & Tools: Pivot Table, Pivot Chart, Slicers, and advanced data visualization techniques to ensure effective and insightful business intelligence reporting.

Project Overview:
Conducted an in-depth Operations Research project aimed at minimizing transportation costs, using the Least Cost Method for the initial solution and the Stepping Stone Method to reach optimal solutions. This project formed part of a critical academic assessment, resulting in a research paper that demonstrated both theoretical understanding and practical application.

Key Responsibilities and Achievements:

• Analyzed transportation models to identify the most cost-effective allocation of resources, reducing total transportation costs by approximately 15% compared to standard allocation methods.

• Utilized data-driven techniques to categorize and optimize distribution routes, ensuring minimal transportation expenses across multiple scenarios.

• Employed POM-QM software to validate manual calculations, enhancing accuracy and saving approximately 40% in processing time.

• Generated insights into cost reduction strategies through data visualization and reporting, presenting findings that highlighted optimal distribution strategies for various transportation needs.
  
  

Impact and Value Added:

• Achieved a measurable reduction in projected transportation costs, demonstrating the effectiveness of data-driven decision-making in operations research.

• Enhanced understanding of advanced optimization techniques among team members, reinforcing practical skills in data analysis and mathematical modeling for real-world applications.