Jurnal Sistem Teknik Industri

The Design of a Demand Forecasting Model of Glass Bottled Tea Products With Machine Learning Approach

2024-12-02T14:08:22+07:00

An accurate sales forecasting is crucial to the profits earned because it affects the company's stock management. With computational support, machine learning and artificial intelligence can continuously and automatically recognize patterns in data, thereby reducing the risk of demand unpredictability. PT XYZ is one of the companies in industrial sector that produces various beverage products. The factory in Medan. One of the products is the tea glass bottle. At PT XYZ, there are frequent differences between forecasting data and sales data, causing high error rates in production planning accuracy. This study aims to analyze the most effective model for forecasting future sales by comparing the accuracy of a Machine Learning-based forecasting model with the existing forecasting method currently employed at PT XYZ. This research was conducted using the Recurrent Neural network (RNN) method as part of the Machine Learning approach. The data that was inputted to the programme was weekly demand data, calendar day off data, temperature data, and population data. The forecasted data is weekly demand. Based on the company's historical data, a demand graph is obtained which has a cyclical pattern. From the results of forecasting using Machine Learning, an accuracy value of 99.47% is obtained with an error rate of 0.53%, which is still below the tolerance limit set by the company. The error rate shows a decrease of 14.72% compared to the error value in the previous company model. This decrease is expected to help control inventory more effectively.

Enhancing Human Resource Performance: An Evaluation Study Using the Human Resource Scorecard Method at PT. Wanxiang Nickel Indonesia in Morowali

2025-01-03T19:58:45+07:00

This study evaluates the performance of human resources at PT. Wanxiang Nickel Indonesia (WNI), is influenced by a lack of responsibility, low awareness of work risks, and insufficient employee skills. The research method employed is the Human Resources Scorecard (HRS), involving a Likert scale for Key Performance Indicators (KPI) and Pairwise Comparison in the Analytical Hierarchy Process (AHP). The results of the study indicate the highest weight is found in the perspective of employees' intellectual competence (1.079) and work competence (0.981), followed by emotional competence (0.303) and personality competence (0.246). This research aims to improve human resource (HR) performance and productivity within the company. The intellectual and work perspectives play a significant role in the evaluation, while emotional and personality perspectives also contribute meaningfully. Recommendations: the company needs to develop these perspectives in a balanced manner to enhance overall human resource performance and productivity at PT. WNI.

Model for Determining the Optimum Marketing Strategy for Schneider Electric Products Using Game Theory Approach (Case Study: Authorized Distributor Schneider Electric)

2025-01-23T09:47:59+07:00

Market competition, sales area, and similar business process between authorized distributors of Schneider Electric Medan namely KAK and PJ have an impact on the problems experienced by consumers, resulting in sales decrease and unachieved targets by distributors. Optimal marketing strategy modeling for each distributor is needed to restore consumer confidence and also increase sales. Game theory is used to analyze the most optimum applications by finding the best mathematical model for each distributor, where each player is expected to find the most optimum indicator in maximizing profits or minimizing losses. The novelty in this study is related to the application of strategy implementation (sub-variables) which will be actions that can be taken in implementing selected strategies (main variables) for each distributor. The results of the study show that the optimum marketing strategy of the main variable for KAK is the payment strategy, and for PJ is the aftersales strategy, with the saddle point value of the game is 17.73. The applications of the optimum marketing strategy for KAK payment strategy are COD, Net OEMD, and CIA, while for PJ aftersales strategy are Warranty, Technician, and Service Center, with the saddle point value of the game is 28.18.

Increasing the Stock Taking Process Accuracy for ISO 9001 Quality Management System Fulfilment at 2W EV Manufacturer, Cikarang, West Java

2024-12-04T10:39:27+07:00

The two-wheeled EV (2W EV) manufacturer at the centre of this study prioritizes efficiency, innovation, and adherence to ISO 9001:2015 quality requirements. However, the company encountered issues with its Supply Chain Management procedures during the ISO Surveillance 1 period, where one of the implementations was an internal audit in line with BPMS. Specifically, the results showed that the physical stock that matched the system was only 80%, which was very far from the KPI target of 95%. It hindered operations and impacted customer satisfaction. Business Process Improvement (BPI) was applied to examine the current stocking system. Comparing current processes with the best practices specified in BPMS and ISO 9001:2015 criteria was the first step in performing a thorough gap analysis. The results showed that human mistakes, machine calibration problems, a lack of standard operating procedures, inconsistent counting techniques, mixed scrap materials, and an unmanaged warehouse environment were causes of stock variations. The company made strategic changes, such as standardizing material labelling to avoid recording errors, substituting material loan for the interdepartmental purchasing system to increase accountability, and updating and enlarging SOPs to guarantee process uniformity. The business also installed an ERP system to enhance stock accuracy, facilitate real-time monitoring, and combine warehouse operations with digital tracking. As a result, by September 2024, the efficiency of stock control had increased by almost 86%. The increased stock control not only improves compliance with ISO 9001 standards, but it also increases overall operational efficiency by reducing production problems and increasing customer satisfaction.

Quality Optimization of Fuel Transportation Tank Production Process Using Design of Experiment (DoE) at PT. Sejahtera Mandiri Pekanbaru

2024-12-20T10:11:54+07:00

With an average error rate of 2.3%, the production of fuel transportation tanks at PT Sejahtera Mandiri Pekanbaru has defects that exceed the 2% tolerance level. Porosity, undercut, and crack are the most frequent types of defects that result in additional cost and rework time. This study aims to improve the quality of the fuel transportation tank production process at PT Sejahtera Mandiri Pekanbaru. Data shows that production errors such as porosity, undercut, and cracks have an impact on the quality of the final product. This study used a 2^k full factorial design to find the main components that affect the quality of welding results. The results show that operator certification and electrode temperature oven parameters have a great influence on manufacturing defects. Maximizing the use of electrode oven at 260°C-425°C for 1 hour, electrode moisture can be eliminated and porosity can be reduced. The DoE approach is good for finding and optimizing production features to reduce defect rates. This study suggests operator training and control of process parameters to improve welding quality.

Design of Corrugated Cardboard Product Delivery Allocation Model by Considering Heterogeneous Fleet and Multi Product

2024-12-12T13:11:38+07:00

Manually planning truck usage by estimating the right fleet selection, as well as considering capacity, delivery zones, and variations in demand, often results in inaccuracies. This inaccuracy has an impact on fleet utilization that is not yet optimal, characterized by low capacity used for each trip and the presence of fleets that are idle at certain times. Apart from that, inappropriate delivery allocation planning and less than optimal fleet utilization also affect delivery timeliness. This research proposes a delivery allocation model to solve the Capacitated Vehicle Routing Problem (CVRP) problem using a Genetic Algorithm (GA) implemented within the Multi-Objective Evolutionary Algorithm (MOEA) Framework. The model is designed to address CVRP by considering heterogeneous vehicle fleets, product variability, and diverse delivery destinations. The chromosome representation in the model describes the sequence of customer visits by the available fleet, while the fitness function is focused on minimizing the total traveled distance in order to maximize the efficiency of vehicle capacity utilization. Based on test results at PT In addition, this model succeeded in reducing the number of fleets used by up to 50%.

Parallel Scheduling using Genetic Algorithm and Knowledge Based Approach

2024-12-12T10:47:16+07:00

Production scheduling are very important considering the complexity of the production system. This study aims to solve parallel machine scheduling to get the best job sequence and minimize lateness. Genetic algorithm is optimization algorithms by implementing evolution process and eliminating bad solutions. Knowledge based approach (KBA) solve problems by creating a computing system to imitates human intelligent behavior. Genetic algorithm and KBA are combined with the earliest due date (EDD) rule to produce an inference engine to build more adaptive population initialization. The results of the proposed scheduling show that the rules successfully guide the search process more adaptively. The genetic operation increasing the fitness value when the job is overload or underload. When the job is underload fitness increases by 3.56%, there is no lateness and load capacity ratio (LCR) increase by 4.67%. When the overload fitness increases by 1%, lateness decreases by 4.57%, and LCR decreases by 7.56%. The increase of fitness value shows better results of the proposed job sequence with minimum lateness. The implementation of integration genetic algorithms and KBA using VB.Net language requires a reasonable computing time, which is an average of 32 seconds when running.