| 000 | 03603ntm a2200349 i 4500 | ||
|---|---|---|---|
| 003 | MY-KuUP | ||
| 005 | 20260727100319.0 | ||
| 006 | t|||||r|||| 000 0 | ||
| 007 | ta | ||
| 008 | 260727t20252025my a|||fr|||| 000 0 eng d | ||
| 020 |
_aTHE0010762 (Local) _qHardback |
||
| 040 |
_aUMPSA _beng _cUMPSA _erda |
||
| 090 | _aFTKKP .S63 2025 r Bc. | ||
| 100 | 0 |
_aSoddiki Sylvester Bin Theobold, , , _eauthor. |
|
| 245 | 1 | 0 |
_aOptimizing risk management strategies for reverse osmosis in pharmaceutical purified water systems through enhanced integration of machine learning and quantitative risk assessment / _cSoddiki Sylvester Bin Theobold |
| 264 | 1 |
_aKuantan, Pahang : _bUMPSA, _c2025 |
|
| 264 | 4 | _c© 2025 | |
| 300 |
_axi, 66 pages : _billustrations ; |
||
| 336 |
_2rdacontent _atext |
||
| 337 |
_2rdamedia _aunmediated |
||
| 338 |
_2rdacarrier _avolume |
||
| 347 |
_2rda _atext file _bPDF |
||
| 500 | _aFaculty of Chemical and Process Engineering Technology | ||
| 502 | _aFinal Year Report (Bachelor of Manufacturing Engineering Technology (Pharmaceutical) with Hons.) -- Universiti Malaysia Pahang Al-Sultan Abdullah - 2025 | ||
| 504 | _aInclude bibliographical reference | ||
| 520 | 3 | _aWith the emergence of the Fourth Industrial Revolution (IR 4.0), more industries, like the pharmaceutical industry, are focusing on digital technologies that can add efficiency and reliability to their process, such as reverse osmosis (RO) in purified water systems. This study investigates the integration of machine learning (ML) with quantitative risk assessment (QRA) as tools to optimize RO processes in pharmaceutical purified water systems. The main objectives of this study were to apply a multivariable mathematical model, integrate QRA with ML techniques, and validate the predictive models with QRA outputs. The study followed a systematic methodology to achieve its objectives. Firstly, critical process parameters (CPP) were identified using the multivariable mathematical model, followed by data simulation with @RISK software. The Friday 13th (Fr 13) framework was implemented using Monte Carlo simulations to identify unexpected failures. 10,000 iterations were generated from @RISK to evaluate the distribution of risk factors. The data was then imported into RapidMiner for ML analysis to create predictive models. The ML approach used Random Forest ML algorithms to improve risk prediction and classification. The dataset was split into training (70%) and testing (30%) subsets to ensure robust model validation. The analysis, which included 10,000 data points and represents 100% of the dataset, showed a failure rate of 20.8% for @RISK and 20.82% for RapidMiner. The results showed a slight difference (0.02%) in the risk of failure, demonstrating it accuracy of the predictive models. The integration of ML improved predictive accuracy, with the Random Forest model identifying risk factors with a high classification performance (97.8%). These results demonstrate the accuracy of the developed machine learning model and its potential for accurate risk assessment in RO systems. To conclude, the integration of ML and QRA provides a reliable and efficient approach to risk assessment in pharmaceutical RO systems. These results highlight the potential of ML to automate risk analysis and optimize operational performance. | |
| 610 | 2 | 0 |
_aFaculty of Chemical and Process Engineering Technology _xDissertations |
| 650 | 0 |
_aUniversities and colleges _xDissertations |
|
| 650 | 0 |
_aTheses _xDissertations |
|
| 856 | _uhttps://umpir.ump.edu.my/id/eprint/47706 | ||
| 942 |
_2lcc _cPSM |
||
| 999 |
_c105304 _d105310 |
||