Module08Video03

Summary

This video explores advanced features in Simio for modeling a hot dog stand, focusing on vendor capacity and customer arrivals. It demonstrates how to adjust vendor availability during peak times, such as lunch rush, by using a work schedule. The video also covers how to implement an external arrival table, importing customer arrival data from Excel, to simulate realistic customer flow. Additionally, it highlights the use of arrival time deviations and no-show probabilities to enhance the model's realism, offering flexibility in experimentation and better resource management.

Takeaways

• 😀 The hot dog stand model uses time-varying arrival rates, with different arrival rates during morning, lunch rush, and afternoon periods.
• 😀 Two vendors are added during the lunch rush to accommodate increased customer demand, improving capacity during peak times.
• 😀 The work schedule for vendors is dynamically adjusted using a 'standard week' with 1 vendor from 9-11 AM, 2 vendors from 11 AM-1 PM, and 1 vendor from 1-3 PM.
• 😀 Simio’s work schedule feature allows for dynamic resource management based on specific time periods throughout the day.
• 😀 Extending the vendor’s working hours to 4 PM prevents customers from staying in the system after the originally scheduled end time of 3 PM.
• 😀 An external arrival table is introduced, allowing customer arrivals to be controlled by real-world data rather than random or time-varying rates.
• 😀 External data (e.g., from an Excel spreadsheet) can be imported into Simio to specify exact customer arrival times.
• 😀 Arrival time deviations can be added, allowing customers to arrive slightly earlier or later based on a specified uniform distribution.
• 😀 The no-show probability feature allows for modeling real-world scenarios where not all scheduled arrivals show up, useful in services like doctor’s offices.
• 😀 The use of an arrival table in Simio allows for experimentation with actual historical arrival data, enabling more realistic simulations.
• 😀 Simio’s 'Write' step can generate external data, which can then be imported into another model, demonstrating a way to use simulated data in future experiments.

Q & A

What is the purpose of adding an additional vendor during the lunch rush in the simulation?

The purpose of adding an additional vendor during the lunch rush is to accommodate the higher volume of customers expected during that time, reducing congestion and improving service efficiency.

How does the work schedule help in managing vendor capacity?

The work schedule allows the system to define the number of vendors available at different times of the day. For example, during the lunch rush, two vendors are scheduled, while one vendor is available during the morning and afternoon periods.

What is the impact of using a fixed vendor capacity compared to a work schedule?

Using a fixed vendor capacity would mean the same number of vendors are available throughout the entire day, regardless of the customer arrival rate. A work schedule adjusts vendor availability based on demand, such as adding extra vendors during busy times like the lunch rush.

How does the arrival table feature improve the simulation model?

The arrival table allows the simulation to use real-world data for customer arrivals, improving accuracy and enabling experiments with actual arrival patterns rather than relying on random or time-varying arrival rates.

What type of data is used in the arrival table, and how is it imported?

The arrival table uses external data, such as arrival times stored in an Excel spreadsheet. This data is imported into the model and bound to a data table, which then controls the arrival of customers in the simulation.

What is the significance of the time deviation feature in the arrival table?

The time deviation feature allows for the introduction of variability in the arrival times, simulating real-world scenarios where customers may arrive earlier or later than scheduled, providing more realistic modeling of customer behavior.

What problem occurs if a customer is still in the system at 3 PM, and how is it addressed?

The problem is that the vendor's work schedule ends at 3 PM, but a customer may still be in the system. This is fixed by extending the service period beyond 3 PM, ensuring that customers are still served if they are in the system when the schedule ends.

What could be a real-world example where the arrival time deviation and no-show probability would be useful?

A real-world example could be a doctor's office or dentist's office, where appointments are scheduled at specific times, but patients might arrive late (time deviation) or not show up at all (no-show probability), making these features useful for simulating appointment scenarios.

What type of external data could be used in the arrival table aside from Excel spreadsheets?

Aside from Excel spreadsheets, external data could come from time studies, observational data, or recorded customer arrival times from previous days or different locations, all of which can be imported into the model for experimentation.

How does the use of the 'write step' in the model help generate external arrival data?

The 'write step' allows the model to generate and write arrival times to an external file during the simulation. This data can then be imported into another model, allowing for the use of real-time or generated data in future simulations.