As Arge Bilişim, we would like to share with you the 'line balancing module' that we have designed to meet a very important need in factories.

First, let's briefly touch on what line balancing is.

Businesses have turned to assembly line production systems to more efficiently use limited production resources and reduce production costs. In these production systems, we refer to the path where the parts of the product are made at each workstation and progress as the parts are transformed into the product as a "line".

Line balancing aims to equalize the speeds of these workstations to maximize machine and labor utilization.

For each operation in the production line to be carried out, the preceding processes must be completed. If you cannot establish a balanced line, bottlenecks will occur within the line, resulting in some operators waiting idle while there may be accumulations in other areas. And this situation will lead to a loss of efficiency. The station with the most accumulations, i.e., the weakest station in the flow, will also determine the capacity of your production line. To give a simple example, let's consider producing a product that goes through a total of 5 different processes. Let's assume each of them has a production capacity of 100 units per hour. Only the 2nd station has a production capacity of 90 units per hour. In this case, if you haven't balanced your line, there will be constant accumulations from the first station to the second, while the operators after that will wait idle. And you will have a production loss of 10 units per hour.

Moreover, in this example, we assumed that the capacities of 4 stations are equal, but in reality, the durations of operations are not the same, and the speed of each operator, their competencies, skills, and the difficulty levels of tasks are different. Therefore, many parameters like these need to be taken into account when setting up the line.

Proper line balancing will ensure effective use of labor and resource capacities, reduction of idle times, and increased efficiency leading to reduced production costs.

Okay, how are businesses currently handling this?

In most existing production systems, while engineers theoretically conduct line balancing studies, operator scheduling cannot be done, or if it is done, it's not effective. This is because operator scheduling is a situation that can vary based on real-time data and is usually carried out by line supervisors who are continuously on the shop floor, rather than by engineers. When an order comes in, the engineer first establishes the operation and standard times for the model. Then, at best, by setting a target efficiency, workloads are calculated, and this list is given to the line supervisor. For example, the information that 1.5 people are needed for the side bending operation is provided by the engineer, but there is no information on which 1.5 people, or how efficiently these individuals have worked in this operation before. Therefore, this information remains purely theoretical. Line supervisors place individuals based on their experience of who can do which job or who has the capacity to do it. If anyone is absent from the line, the balance of the entire line will be disrupted, and they will have to rethink the entire setup from scratch. Redesigning is difficult and often not feasible as it leads to time loss and slows down workflow.

Without a system like ArgeMAS that measures productivity and quality, neither the standard time for the product nor individual productivity will be taken into account. Even the loss of efficiency between them may go unnoticed.

This complex decision, which cannot be achieved by observation alone, is too important to be left to individual initiative.

In systems where manual labor is intensive and variability is high, the efforts made so far regarding line balancing have been far from solving the problem completely.

Okay, how did we solve this fundamental problem that affects the future of factories?

Combining practical knowledge with theoretical knowledge in production will give much better results. Therefore, there is a need for a structure that combines real-time data collected from the field, accumulated past knowledge, and current engineering studies. With this aim in mind, Arge Bilişim engineers have developed a mathematical and specialized assignment optimization algorithm to simultaneously solve optimization problems in line setup and operator scheduling.

Firstly, with the ArgeMAS system we have implemented in factories, we can access real-time data on each operator's operational efficiency, quality, downtime, and other factors on an operational basis. We also know the standard times for each operation with the created worksheets. However, not every model and operation may be suitable for the capabilities of the line. To overcome this, we control the ability of operators to perform each operation using two complementary methods.

The first method is operation similarity classification. When defining operations, we classify them based on their methods, grouping together operations with similar execution methods. For example, the punching operation, attaching components operation, and quality control operation have completely different execution methods. While the punching operation requires alignment and rhythm skills, the component attachment operation requires alignment and finger skills. From this, it can be inferred that someone who efficiently performs the punching operation can easily learn the component attachment operation, which requires similar skills.

The other method we use to control operators' ability to perform operations is the difficulty levels of operations. With the ArgeMAS system, difficulty criteria are determined, allowing the difficulty levels of each operation to be identified. For example, criteria such as long training time, high physical strain, and high risk of quality errors are some of the criteria that make an operation difficult. By expanding on such criteria, the difficulty levels of operations are determined using analytical job evaluation methods. Taking into account the similarities between operations and their difficulties, the Arge Bilişim assignment optimization algorithm can easily predict the ability of an operator who is successful in one operation to perform another operation.

In fact, this is exactly what line supervisors on the production floor try to do, but since each line supervisor can only make evaluations based on their own experience, the margin of error is unfortunately very high. While benefiting from the managerial skills of line supervisors in the field, the decision on how to ideally set up the line is entirely made by the Arge Bilişim assignment optimization algorithm, which is a completely real engineering effort.

With the Arge Bilişim assignment optimization algorithm:

You can easily access information such as how much time your operators will spend on each operation, what percentage of your operations will be done by which operator, and obtain detailed reports based on quantity or bundle. Here, the efficiency value of each operation, operator, and production line is planned.

Taking into account the workload of operations, the model, operations, similarity groups, and difficulty levels are considered to maximize line efficiency by the algorithm. The assignment made as a result of this allocation ensures that the efficiency of the resulting line is calculated entirely based on real data. Instead of static capacity, dynamically calculated capacity based on real data is used as an important piece of information in production planning. Production planning based on completely real data replaces production planning made in the form of "If we produce an average of 1000 pieces per day, we will meet this order by Friday". This also prevents delays in deadlines due to erroneous capacity planning.

Moreover, if there is any change in the field, the line setup will need to be changed and a new ideal structure found. For example, if 3 operators do not show up for work on a given day, this situation can cause complete chaos in the field, but the Arge Bilişim assignment optimization algorithm quickly and effectively replans the most ideal line structure based on these changes and presents the real plan to the user again.

The ideal line structures where models will be produced most efficiently are also determined by the Arge Bilişim assignment optimization algorithm. For example, if there are 15 lines in your facility and you want to plan 20 models on these lines, the system treats the facility as a single line, and the most ideal 15 lines where these 20 models will be produced are recreated. Thus, an increase in efficiency can be achieved by solving the bottleneck operation on one line with a competent but idle operator found on another line, compared to lines set up without the program.

To summarize the results of the Arge Bilişim assignment optimization algorithm:

Firstly, ArgeMAS measures efficiency and quality. The "Arge Bilişim assignment optimization algorithm" then establishes the most efficient and balanced lines using real operations and operators based on these efficiency and quality results.

Secondly, it provides real production capacity information for production planning.

Another third benefit is that it reduces the workload of line supervisors, allowing them to focus more on managerial tasks.

Another benefit is that the time for line setup, shortened with the bundle system, will be further reduced with this algorithm. During model transitions, the line supervisor will not experience uncertainty about which operator to assign to which operation. They can easily organize their lines with the reports they have.

Fourthly, reconfiguring the line to adapt to production variations is quite difficult. With this algorithm, establishing the most ideal line according to current conditions will be easy and fast.

Planning multiple models on multiple lines can also be considered as another benefit, as it creates the most ideal line structures.

We can also use it in many other areas that directly support efficiency increas

With the Arge Bilişim assignment and optimization algorithm, you can prevent efficiency losses and reveal your true potential in production.

 

• Enabling real-time measurement of productivity and quality
• Increasing efficiency with regular workflow
• Facilitating the implementation of a productivity and quality-based bonus system
• Enabling the promotion of productivity and quality through a bonus system
• Facilitating product flow and optimizing unnecessary movements
• Reducing production costs by decreasing indirect labor
• Making operation times accurate and predictable
• Establishing controlled and efficiency-enhancing intermediate stocks
• Minimizing downtime
• Facilitating line balancing
• Preventing operators from idle waiting
• Facilitating process controls
• Enabling operators to perform multiple operations
• Enabling operators to use multiple machines
• Implementing an improvable production system
• Ensuring correct line structuring
• Establishing efficiency-focused factory management
• Developing management, production, and quality philosophies
• Implementing an objective management and salary system
• Ensuring accurate product pricing