Scribd
Upload
15 views2 pages

IPE 512 Assignment Report Davis Isibor

The report details an assignment by Davis Isibor on forecasting software releases using a Genetic Algorithm during his industrial training at Digital Tech Firm. The model predicts software releases for five months, aiming to minimize forecasting error while maximizing revenue, resulting in a total forecasted revenue of ₦10,600,000. The assignment enhanced his understanding of optimization algorithms and their application in business scenarios.

Uploaded by

DAVIS ISIBOR
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
15 views2 pages

IPE 512 Assignment Report Davis Isibor

The report details an assignment by Davis Isibor on forecasting software releases using a Genetic Algorithm during his industrial training at Digital Tech Firm. The model predicts software releases for five months, aiming to minimize forecasting error while maximizing revenue, resulting in a total forecasted revenue of ₦10,600,000. The assignment enhanced his understanding of optimization algorithms and their application in business scenarios.

Uploaded by

DAVIS ISIBOR
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

IPE 512 Assignment Report

Name: Davis Isibor

Reg No: 20

Course: IPE 512 – Advanced Modeling Techniques

Lecturer: Dr. Ezeliora

IT Company: Digital Tech Firm (Software Products)

Model Used: Genetic Algorithm (Reg No 11–20 Group)

Tool Used: Python

1. Introduction
During my Industrial Training at a digital technology company, I worked on several product
cycles involving the release of web platforms, APIs, and software tools. The company’s
operations followed agile workflows where product releases occurred monthly. This
modeling task simulates and forecasts the number of software releases over a future five-
month period using the Genetic Algorithm technique.

2. Problem Statement
The objective was to forecast the number of software products expected to be released over
the next five months using a Genetic Algorithm. The model also optimized for revenue,
based on the assumption that each product yields ₦200,000. The result aimed to minimize
forecasting error while maximizing revenue potential.

3. Methodology – Genetic Algorithm


Genetic Algorithms are bio-inspired optimization techniques based on the principle of
natural selection. The model simulates evolution by generating a population of forecasts,
evaluating their fitness, selecting the best, combining them (crossover), and applying
random variation (mutation). The steps followed were:
1. Generate 10 random forecast sets
2. Define a fitness function balancing MSE and total revenue
3. Select best performers and evolve over 10 generations
4. Output the best forecast and associated revenue
4. Assumptions
Historical monthly release data (Jan–Oct): [2, 3, 4, 3, 5, 4, 6, 7, 8, 9]
Forecast period: Nov 2024 – Mar 2025
Revenue per product: ₦200,000
Forecast evaluation metric: Mean Squared Error (MSE)

5. Final Forecast Output


Nov 2024 10
Dec 2024 9
Jan 2025 11
Feb 2025 10
Mar 2025 13
Total Revenue (₦) ₦10,600,000
Fitness (MSE): 0.86

6. Reflection and Learning


This assignment provided me with practical experience applying optimization algorithms to
real-world business scenarios. I learned how to balance forecasting accuracy with business
goals (like revenue), and how to implement a Genetic Algorithm in Python. This deepened
my understanding of evolutionary computing in engineering contexts.

7. Tools Used
• Language: Python 3
• Libraries: NumPy, Pandas
• Environment: VS Code

You might also like