new program structure

MMC.py, MMC_OO.py

M/M/c (multiple server queue model. This demonstrates both the multiple capacity Resource class and the observe method of the Monitor class. Random arrivals, exponential service-times. (TV)

bcc.py, bcc_OO.py

Determine the probability of rejection of random arrivals to a 2-server system with different service-time distributions. No queues allowed, blocked customers are rejected (BCC). Distributions are Erlang, exponential, and hyperexponential. The theoretical probability is also calculated. (TV)

callCenter.py, callCenter_OO.py

A call center with agents with different skills. The model returns the frequency distribution of client waiting times, the percentage of reneging clients, and the load on the agents. This model demonstrates the use of yield get with a filter function. (KGM)

cellphone.py, cellphone_OO.py

A cellphone cell with blocking (that is BCC discipline). The program simulates the operation for 10 observation periods and measures the mean and variance of the total time blockedr and the number of times blocking occur in each hour. An observational gap occurs between the observation periods to make each one's measurement independent. (TV)

centralserver.py, centralserver_OO.py

A primitive central-server model with a single CPU and a single disk. A fixed number of users send "tasks" to the system which are processed and sent back to the user who then thinks for a time before sending a task back. Service times are exponential. This system can be solved analytically. Trace statements have been removed. (TV)

jacksonnetwork.py, jacksonnetwork_OO.py

A Jackson network with 3 nodes, exponential service times and probability switching. The simulation measures the delay for jobs moving through the system. (TV)

Miscellaneous Models

bank08renege.py, bank08renege_OO.py (partial conversion)

Use of reneging (compound yield request) based on bank08.py of the tutorial TheBank. Customers leave if they lose patience with waiting.

Carwash.py, Carwash_OO.py

Using a Store object for implementing master/slave cooperation between processes. Scenario is a carwash installation with multiple machines. Two model implementations are shown, one with the carwash as master in the cooperation, and the other with the car as master.

CellularAutomata.py

A two-dimensional cellular automaton. Does the game of Life. (KGM)

demoSimPyEvents.py

Demo of the event signalling constructs introduced in SimPy 1.5. Three small simulations are included: Pavlov's drooling dogs, an activity simulation where a job is completed after a number of parallel activities, and the simulation of a US-style 4-way stop intersection.

shortestPath_SimPy.py, shortestPath_SimPy_OO.py

A fun example of using SimPy for non-queuing work. It simulates a searcher through a graph, seeking the shortest path. (KGM)

Machineshop.py, Machineshop_OO.py

An example of the use of the interrupt() method. It simulates a workshop with n machines which break down and one or more repairmen which have to repair them. (KGM)

Market.py, Market_OO.py

A supermarket checkout with multiple counters and extended Monitor objects. Written and analysed by David Mertz in an article for developerWorks (). (MM)

Movie_renege.py, Movie_renege_OO.py

Use of reneging (compound yield request) constructs for reneging at occurrence of an event. Scenario is a movie ticket counter with a limited number of tickets for three movies.

needResources.py, needResources_OO.py

Demo of waitUntil capability introduced in SimPy 1.5. It simulates three workers each requiring a set of tools to do their jobs. Tools are shared, scarce resources for which they compete.

SimPy_worker_extend.py, SimPy_worker_extend_OO.py

Factory making widgets with queues for machines. (MM)

WidgetPacking.py, WidgetPacking_OO.py

Using buffers for producer/consumer scenarios. Scenario is a group of widget producing machines and a widget packer, synchronized via a buffer. Two models are shown: the first uses a Level for buffering non-distinguishable items (widgets), and the second a Store for distinguishable items (widgets of different weight).