1 Introduction

1.1 What is iTS MAIT?

The innovative Transport Simulator by MAIT international e.V. is an easy-to-use micro-simulator software, intended to simulate and evaluate a set of ``innovative transportation systems''. The software has been developed according to the MAIT system specifications, and software specifications, presented at the MAIT Meeting 2001, Loughborough, UK. its-0.51 also satisfies the specifications for the PRT simulation software articulated by the EUNITRANS group during the Automated People Mover conference APM'99, Copenhagen, Denmark.

Even though the ultimate intention for this software is the simulation of a complete MAIT system, its current use is focused on PRT or AGV systems.

Essentially, the software predicts performance and cost figures of a specified system. The input data will include the network topology and assumed demand-patterns, maximum accelerations, line-speed, boarding-times, component-costs, etc. Output data will include waiting times, journey times, running costs and capital costs.

The performance-relevant technologies implemented are

• a novel version of a vehicle follower control algorithm. This control system is thought to allow close-to-physically achievable results (see also the discussion under ``Carriers'' in Sec. 5.1).
• Acceleration at stops, diverge and merge behavior.
• Boarding behavior at stops. Currently are two modes implemented:
  • Synchronous mode, where the stop is divided into a loading and unloading area.
  • Asynchronous mode, where users can queue up at all berth of the stop and enter the next available vehicle.

• Shortest estimated travel-time routing. (See Sec. 1.3 under logistics.) Considerable higher throughputs in high-density traffic operation are expected from a global logistic operation system, which is not yet implemented.

The software has been designed to obtain a holistic view of the simulated transportation system. For this reason it covers (or will cover) a variety of system aspects such as:

• Network design and planning, i.e traffic flow/ traveling speed/ waiting times etc. for a given network layout and origin-destination demand patterns. (See scheme of Fig. 1.)

• Cost- and performance analysis i.e prediction of passenger km, vehicle usage, expected investment and operating&maintenance costs, economy of scale, etc.

• Support for technological development of vehicles and infrastructure i.e to incorporate upgrades to modularization and interfaces, control systems, server-networks, logistics, management systems, etc.

• Support for implementation, test and integration, so that simulated hardware modules, such as vehicles and track, can be replaced by their real physical systems, while the logistics/management modules of the software can continue to control the real network-- with almost no modifications.

• Support for geometrical design, study of visual impact and promotion i.e to provide 3D output of simulated network.A simple, but realtime 3D-viewer has been implemented.

For currently implemented features see Sec 1.3.

Figure 1: Network design cycle.

1.2 Who should use iTS MAIT?

Because the simulator covers a large number of development and implementation issues, it is of interest to:

• system developers,
• manufacturers,
• transport experts, consultants, town planners, architects,
• potential operators and
• anybody who dreams of a better way to transport people and goods.

1.3 Currently implemented features

• Easy to use transportation network editor: Track elements, vehicles and users can be selected from a library and placed and edited on the canvas with a graphical 2D network representation. Scanned maps of the target region can be used as background images. The number of different network components (track elements, vehicle-types, users) is still limited but is expected to grow rapidly. Anyway, new track elements can be easily created by copying and editing library text-files.

• Vehicle dynamics: The vehicles are using a vehicle-follower control algorithm. Speed-limits, maximum comfort/emergency/safety acceleration, brake actuation time are respected. Jerk is not simulated explicitly, but jerk limits can be introduced implicitly such that the average vehicle distances are represented correctly--as if jerks were simulated. All parameters can be modified via GUI interfaces (in this case called ``control-panels''). For more information on control issues, see Sec. 5.1 under carriers and Sec. 5.3.

• Passenger-behavior: Currently there are two types of passengers implemented:
  • generic user: this user is making one trip during the simulation, where the origin and destination stop are indirectly determined by a origin-to-destination matrix. See Sec. 4.1 on how to use this feature.

  • test-driver: This user makes one random trip after another. He has been mainly created for test purposes.

• Logistics: Currently three management modules are implemented:
  • Passenger management, which is the only one that interfaces directly with passengers.
  • Carrier management, allocates a vehicle to the passenger management. Optimized empty vehicle management.

  • Track management, knows the network topology, instructs all diverge points to direct the vehicle on the fastest way to the desired destination.
  For more information, see management objects in Sec. 5.1 and Sec. 5.2. .

• Analysis and validation: The current data about each module, i.e vehicles, users, track, managements, can be displayed via control-panels. The data contains module-dependent information about
  • Performance, e.g throughput, average speed, waiting times, traveled passenger km, etc.
  • Economics, e.g initial investment, annual costs, trip costs, etc. and Importantly: Economy of scale. Costs, that depend on the quantity (or length) of modules can be edited in a quantity/prices table. The software will then automatically compute the price, dependent on the size of the network, number of vehicles, etc.
  • important parameters, such as accelerations, line-speed, etc. Most of these parameters can be interactively changed.

• Export results: the most significant data of the current simulation, i.e parameters, performance and costs, can be exported as a tab-separated text file, which is easily imported into all spread-sheet applications such as Excel, kspread, star-office or Gnumeric.

• Save and load: the current state of the simulated network, including vehicles, users and managements, can be saved at any time into a simulation file. This simulation file can be reloaded and the simulation continued from the state when the network was saved.

• Command-line mode: the simulation of a previously edited and saved network can be launched also without graphics. In the Command-line mode simulation times can be reduced considerably. It also allows to run the simulation within scripts or as a batch process. Read more in Sec 4.3.

• Scripting: the simulation of a previously edited and saved network can be used in scripts. Parameterized vehicles and passengers can be added within the script and results can be saved automatically, see Sec 4.2.