When we make proposals for material handling systems to customers, specialized engineers consider proposals from various perspectives so that we can provide the most optimal system. One of the techniques used in this case is system simulation. Using dedicated material handling simulation software, models are created on a computer of the planned system and the existing system. Various conditions, such as the volume of equipment and specifications, are set and simulations are conducted. This makes it possible to assess the movements of the systems and to assess and identify issues for the models created. This information can then be used when considering system optimization. Simulation is also a valuable tool for customers, as the information gained can be used as a reference point for considering appropriateness of introducing facilities and the return on investment. In this section, we will provide an overview of simulation, one of Daifuku's strengths, as well as an introduction of some characteristics and examples of simulation.
Simulation consists of establishing models of actual systems and conducting experiments using these models for the purpose of understanding system behavior and assessing various measures relating to operation. Simulation is one of the most powerful analysis tools used for the verification of the design and operation of complex processes and systems, and has been recognized as an extremely useful tool in each stage of planning, design, and management in various fields.
For our material handling systems, we use simulations in engineering and system design as soon as possible so that we can provide trusted systems to customers. In particular, systems have become increasingly complex in recent years, such as the adoption of high-speed vehicle system STV with two carts on one trajectory and dual-type systems for stacker cranes (Photo 1). It has become even more difficult to assess the capacity and performance of such systems, with an increasing number of cases in which advance verification is necessary. In addition, customers now have needs related to risk management, such as the need to assess system issues and prepare countermeasures before equipment is introduced.
1．Determining the Efficiency of Systems
At Daifuku, we use simulations to assess whether facilities fulfill the conveyance capacity required by customers before they are introduced. We consider whether each piece of equipment functions efficiently and verify system performance with factors, such as conveyance volume and lead time, as performance indicators.
For example, for a system like that displayed in Figure 1, a system is composed of a unit load AS/RS, STV, picking station, and storage and retrieval station. The basic flow of goods consists of the following three flows:
- 1）Storage cycle
Storage station→ STV→ AS/RS
- 2）Picking cycle
AS/RS→ STV→ picking station→ picking→ STV→ AS/RS
- 3）Retrieval cycle
AS/RS→ STV→ retrieval station
The purpose of simulation in this model is to consider factors, including equipment specifications, control, and operations, so that each equipment functions effectively and to conduct a verification on whether the required throughput can be satisfied. Main output items include the AS/RS operational rate, the load factor for picking workers, and the number of picking process pallets. The STV conveyance volume (Figure 2) and the results and assessment details in how the required throughput not being reached, as well as some examples of countermeasures will serve as an example.
Despite how the operational rate of the AS/RS is low and there is room to increase retrieval quantity, the required quantity is not achieved.
Due to insufficient STV performance, the retrieval aisle conveyor is always nearly full, and retrieval is not possible. → Increase the number of STVs.
Due to the insufficient number of aisle conveyor buffers on the retrieval side of the AS/RS, it is not possible for the AS/RS to keep up with the dispersion in conveyance tack from the STV. → Increase the number of aisle conveyor buffers.
Due to insufficient picking station capacity, retrieving requests are not supported. → Increase the number of picking workers.
As indicated in the example, by conducting a simulation it is possible to verify whether the required throughput is being fulfilled. There could be a single factor or multiple factors causing a drop in overall efficiency, and simulations can be used to pursue the causes. In addition, by conducting trials on various cases to determine what improvement measures should be adopted, the effects can be determined.
In this manner, we use simulations to assess the effects of equipment during the system planning stage and consider what can be done to improve the efficiency of systems by going through a process that includes the assessment of bottlenecks, the pursuit of causes, and improvements in response to issues.
2．Characteristics of Simulations
Improving the Accuracy of Simulations
Consistency between models and actual equipment is an important point in simulations. At Daifuku, we consider hard numbers, including the speed and and deceleration speed of equipment; evaluate intangible factors such as picking work, worksite operations and the incorporation of actual equipment control; and conduct field surveys after the introduction of equipment in order to compare simulations and actual equipment (Figure 3). As a result, we have received favorable evaluations from third-party organization and have received certification of the validity of our models.
Consideration of Layout and Logic for Each Individual System
Another characteristic of simulations is the consideration given to the layout and logic for each individual system. A material handling system is a custom-made product that reflects the needs of each customer.
In addition to our standard logic, we consider factors such as the layout and logic specific for each individual system and use this information as feedback for system construction (Figure 4).
Use in Product Development As Well
Another characteristic is the use of simulation when we develop new models. We conduct simulations and use the issues identified as feedback for the models being developed. This provides Improvement measures in the control of actual machines. In this manner, we work to make products more effective and reduce worker-hours for internal trials and tests. By shortening the development period, we make it possible to supply products more promptly.
3．Examples of Simulation
The first example is a distribution center for a major overseas supermarket (Diagram 5).
This center contains a system in which pallets are retrieved from an AS/RS, depalletized by case, and then automatically shipped by roll box pallet separated by item. In the simulation, in addition to determining proportions for the entire system, the RM specs and number of STV units were calculated to confirm the specifications for the equipment. Furthermore, in the process of conducting the simulation, it was discovered that a bottleneck was caused by the case turning equipment, although the throughput was sufficient as single component. Based on this discovery, the equipment performance was improved and the storage amount before and after devices was increased.
By incorporating these considerations in system planning, it was possible to respond smoothly to full operations during the Christmas season when there is a high load on the system.
The second example is a transport system in a production process (Figure 6).
For transport in the production process, in addition to the transport volume, the lead time required for transporting to manufacturing equipment is also important. This is because delays in supply to manufacturing equipment can lead to system suspension and have an impact on the throughput to customers.
The transport system in this example is a system that supplies raw materials to manufacturing equipment from an AS/RS, collects products after processing is completed, and delivers them to the AS/RS once again. Typically, many buffers have to be secured to ensure that manufacturing equipment are not suspended. We conducted a test assuming difficult conditions, such as concentrated transport requests, and changed the number of buffers to verify whether the lead time could be met (Figure 7).
Furthermore, we frequently receive request to do simulations for conveyor equipment in manufacturing after installation, such as to investigate the impact caused by an increase in production volume. Once a simulation model has been made, we can promptly support such requests in the future.
Being able to respond to the needs of customers in this manner is one of Daifuku's strengths.
Movements of People
The third example is a simulation of the movements of people in a hospital for a report delivered to the Healthcare Sector of the American Society for Computer Simulation (Figure 8).
This example involves the reconstruction of a ward at a hospital in which the layout would be changed from the current ward to the new ward. The issues were how the lines of visiting patients would change as a result and what could be done to reduce waiting times for treatment. We developed a model of the future layout based on data for current conditions, including treatment time and time at the payment counter, and we conducted an analysis on waiting time reduction from the perspective of both patients and staff members. In order to reduce waiting time for patients, we measured of the effect of changes in the arrival intervals of patients with the introduction a reservation system and considered of the number of service counters.
Research on the efficiency of hospitals to reduce waiting time by patients has been the focus of much attention recently in the simulation field.
The benefits of simulation include the following:
- 1. It is possible to forecast the operation and performance of systems in advance
- 2. It is possible to assess and pursue factors that will have an impact on systems
- 3. It is possible to handle complex systems
By using sumulations, we consider whether the proposed system is the optimal system desired by the customer and verify the impact of a variety of fluctuating variables in advance. In addition, these verification results can be helpful in terms of risk avoidance and the management decision-making process.
Going forward we will work to respond to the needs of customers while making technical developments aimed at faster and more accurate simulations.
Authored by C.D. Pegden, R.E. Shannon, R.P. Sadowski; translated by Soemon Takakuwa
「Introduction to Simulation Using Siman」 (Corona Publishing, 1993)
From DAIFUKU NEWS No.188 (July 2008)