Getting Started - Desktop

Overview

Infrastructure Deterioration Modelling (IDM) is a complex topic. In a modern IDM system, there are a lot of moving parts. To get a working version of Juno Cassandra on your desktop, you need to:

  • Install the Microsoft .Net Core framework, specifically Version 9.0 (free to install)
  • Download and install the latest version of Cassandra Desktop (link will be sent via email)

However, installing the software is simply the starting point. To actually use Cassandra for deterioration modelling, there is a sequence of steps you need to understand and master. These steps include:

  • How to prepare your input file and link this to the Work Bench.
  • How to prepare your Committed Treatments file and link this to the Work Bench.
  • How to set up your workbench file which directs downstream data load operations.
  • How to set up your model configuration file which holds treatment types, unit rates, budget constraints, optimisation weights as well as lookup tables with constants/thresholds required by your model.
  • How to modify lookup table thresholds to tune and direct your model outcome
  • How to inspect and debug your domain model to find and fix anomalies or errors

Without adequate training that helps you understand and practice the above steps, there is no point in even installing the software. To succeed in using Cassandra, you need to commit to a learning path that is long and, unfortunately, may at times be quite intimidating, even frustrating.

How to Get Started

Given the above “warning” about the path to success in working with Cassandra, our procedure for helping users get proficient in Juno Cassandra is as follows:

Step 1: Contact Lonrix

The starting point for your journey in becoming a master-modeller with Juno Cassandra is to reach out to Lonrix via email and let us know that you want to go through the learning path to using Juno Cassandra. If you are not an existing Lonrix client, please provide us also with some background information (what type of infrastructure you want to model, size of your network, etc)

Step 2: Meet with Us

Once we received your email, we will get in touch to set up a meeting with you and your team. In this meeting, we will explain what our training programme for Juno Cassandra involves and clarify any costs involved as far as the training and licensing is involved.

Training and support for Juno Cassandra is not free of charge. As far as licensing is concerned: if you are an existing user of Juno AMS and you want to use Cassandra on one of your licensed networks, there is no extra license fee involved. If you are not a Juno AMS user, or if you want to use Cassandra on another network that is not covered by your existing Juno license, then we can discuss the details in our meeting and therafter provide a quotation for any license fee that may apply.

Step 3: Commit to the Training Programme

If you accept our quotation for training and support, you should be ready to commit to our training programme. This training programme will generally involve several online sessions followed by interactive Q&A sessions.

Please Note: training will take place in sessions staged over several weeks. Many of the concepts involved are relatively complex and will require dedicated focus and committment from you to practice working through the tutorials.

Note

Because the majority of Juno AMS and Cassandra users deal with Road Networks, our standard training is based on, and utilises, our Default Road Network model. We do not currently have pre-programmed models for other infrastructure such as Water Pipeline Networks and Bridge Networks - if you are interested in collaborating with Lonrix to develop such a model for the wider industry, please get in touch!

Training Modules

Module 1 - Software Use

Overview

In this first phase of training, we will focus primarily on how to use the software. You will be provided with a complete set of model input data and setup files so that you can interactively work through all of the steps involved in a model run.

As indicated above, we will be useing the Default Road Network model for all aspects of this training module. However, in this module we will not cover the Domain Model (i.e. the Default Road Network model) in depth. Instead, we will first lay the foundations to using the software and linking your Domain Model to the software.

There are no pre-requisites for this module except for IT permissions and assistance for installing the necessary software and allowing connecting to the Juno Model API. See footnote for details.

Since the tutorial data will involve fewer than 500 elements, Trainees do not require a license to work through this module. This means that anyone interested in exploring Juno Cassandra can participate in this course regardless of whether they are an existing user of Juno AMS.

Topics Covered

  • Installing .NET Core 9.0 and Juno Cassandra (to be completed before first session)
  • Licensing options and installing a license
  • Juno Cassandra Overview - features and options to configure
  • Project folder structure
  • Work Bench setup
  • Model Configuration setups
  • Overview of the the Input File (specific to Default Road Network model)
  • Committed Treatments definition
  • Model Setup File (Treatments, Lookup Tables, Optimisation Weights, etc)
  • Budget Constraint Definitions
  • Defining a Key Performance Indicator (KPI) setup
  • Your first model run
  • Understanding model output files - condition, treatments, spending
  • Post-Processing Tools in Juno Cassandra Desktop
  • Analysing predicted future condition
  • Analysing treatment quantities and spending
  • Analysing budget utilisation
  • Debugging Tools in Juno Cassandra Desktop
  • Troubleshooting and debugging a model
  • Modifying Budgets and re-running your model
  • Modifying Model Thresholds and re-running your model
  • Defining an Objective Function for a Goal-Seek model
  • Running a Goal-Seek Model
  • Budget Scenario analysis
  • Discussion of Next Steps…

Skills Gained

After finishing this module, you should be able to:

  • Understand model configuration attributes and what they mean
  • Set up a folder structure for a model run
  • Correctly structure input, setup and committed treatment files
  • Set up a Work Bench file with one or more model configurations
  • Run a Juno Cassandra Default Road Network model
  • View model outputs using Cassandra Desktop post-processing tools
  • Do basic debugging of a model to find anomalies

What is Not Covered

As discussed elsewhere in this documentation there are two main optimisation types used in Juno Cassandra. One is the Multi-Criteria Decision Analysis (MCDA) method, and the other is the Benefit-Cost Analysis (BCA) method. In this module, we will focus only on the MCDA method, which is a highly practical and flexible model. The BCA model is not covered in this module, but will be covered by future training modules (not yet available).

The above aspects relate mainly to the use of the software itself. As noted above, this module does not provide an in-depth discussion of the Domain model itself, which in this case is the Default Road Network model. However, we will provide an overview of the key features of the Default Road Network model so that you can explore and use the model with some confidence in a training context. This information will be provided at the end of this module as a lead-in to the Module 2 which deals specifically with the Default Road Network model.

Module 2: Default Road Network model

Overview

In this module, we will focus in detail on the Juno Cassandra Default Road Network model.
As with the first module, we will focus on the use of the Default Road Network model in the context of the Multi-Criteria Decision Analysis (MCDA) approach.

Topics Covered

  • Overview of the Default Road Network model
  • The S-curve sub-model for predicting distress development
  • Inverse distribution model for predicting Rutting and Roughness
  • Condition Indexes and how they are calclated - PDI and SDI
  • Treatment Candidate Selection and how to control it
  • Treatment Types and how they are scored for optimisation
  • Input File - format and columns/data required and what they mean
  • Model Parameters - what is being predicted into the future
  • MCDA Setup - optimisation weights
  • Lookup Values - Project Specific values and how they influence your model
  • Lookup Values - Candidate Selection threshold and constants
  • Lookup Values - MCDA Settings and how they influence treatment optimisation

Skills Gained

After finishing this module, you should be able to:

  • Prepare setup files and run the Cassandra Default Road Network model for your own network, provided you are able to prepare the input set yourself.

  • Direct and refine the model by controlling budget constraints, thresholds that determine treatment selection and constants that influence treatment optimisation.

  • All skills from Module 1 should now be re-inforced and available for you to use to compile a modelling report.

What is Not Covered

After finishing Modules 1 and 2, you will have a good understanding of the Default Road Network Model. As indicated above, you will also know how to modify the model to control treatment selection, treatment optimisation and budget constraints at a basic level.

However, you will NOT yet be able to dive into the Default Road Network model engine and make lower level changes. For example, if you want to add a new treatment type that is not part of the Default model, you will not be able to do that before you finish Module 3.

The preparation of the model input file requires analysis of your existing network condition data to join available data statistics into each of the required input columns. This preparation of the model input file is NOT covered in this module.

As with Module 1, in this module we will focus exclusively on teaching you how to use the MCDA method for treatment selection. The BCA model is not covered in this module, but will be covered by future training modules (not yet available).

Note

The preparation of the model input file can be significantly simplified if your road network data is held in a structured database within your Asset Management System. In particular, if your road network data is stored in the Juno AMS, you can use the Data Join Feature to extract and collate data for each network element that you want to include in your model.

However, even if you have the benefit of a tool such as Juno AMS, you should understand that preparation of the model input file is often the most time-consuming aspect of a modelling exercise. When we do modelling on behalf of clients, we use an extensive suite of Data Join tools and custom R-scripts to prepare the input set. This is a specialist task that is not covered in this course. You should consider adding these skills to your repertoire on your journey to becoming an expert modeller.

Module 3: Mentoring Stage

Overview

Once you have completed Modules 1 & 2, you should be ready to undertake a deterioraion modelling analysis using the Default Road Network model on your own network.

Our experience has shown that your first few modelling analyses will invariably lead to situations that you have not yet encountered during your formal training in Modules 1 and 2. For this reason, we offer Mentoring Module in which we provide support for you on an hourly rate basis, thereby helping you to move through your first one or two modelling experiences with confidence.

Support Provided

During the mentorship stage, we can assist you via scheduled online meetings. Typical situations where you may need our help include:

  • You need help preparing your model input file

  • You get an error you do not understand (we can look into it and find the source)

  • You have forgotten something you learnt in Modules 1 & 2 and want a quick 1-hour refresher course

  • Your model does not give expected outputs (we can help find the cause and try to fix it)

  • You do not understand why the model is triggering or not triggering treatments on certain sections (we can help debug into your model and find the causes)

  • You need help compiling your modelling report

Cost of Support

For 2025, our hourly rate for providing mentoring support is NZ$ 180/hour, with a minimum time per interaction of 30 minutes. For your first modelling project, you should budget for 3 to 10 hours of support.

Limitations of Support

While we will endeavour to provide fast support during the mentoring stage when you need it, we cannot be available 24/7. We will do our best to respond within a 24 hour time. If possible, we will set up a online meeting within an hour or two from the time when we receive your distress signal. However, we cannot guarantee that we will be available whenever you need us.

You should therefore plan your first modelling project in such a way that (a) you notify us when you plan to start your project so that we can ensure we are available (and not, for example, take a holiday during that time!); and (b) you provide enough time for yourself so that, in the event that we are not immediately available to support you, you can meet your deadlines even if our support takes 24 hours to ‘arrive’.

Module 4: Modifying your Own Domain Model

Overview

As noted above, in Modules 1 and 2, you will be working with the Default Road Network model developed by Lonrix Limited in association with other industry experts. While the Default Road Network Model is a sophisticated model, as your experience grows, you will almost certainly want to customise the model for the needs of your organisation or network.

You may also want to develop your own domain model for something other than a Road Network, such as a Water Pipeline network or a Bridge Management System.

The Cassandra framework allows you to take full control of your Domain model. A domain model consists of a C# Project managed in Microsoft Visual Studio. The Default Road Network model, for example, is an open-source project that you can download, study and then modify to meet your own needs.

While we understand that computer programming is intimidating for many Civil Engineers, we believe the use of a free, popular and established computer language such as C# is the most flexible and transferable way to develop highly sophisticated domain models to suit your needs. There is an abundance of free online resources to help you develop the experience in C# coding that you need for maintaining a Cassandra domain model.

Module 4 of our training programme is designed to help engineers with an aptitude for analytics and programming get started with coding and maintaining their own Cassandra Domain Models.

Warning

While we will do our best to transfer the necessary skills and knowledge to you in this Module, it should be noted that learning how to code in C# is not for everyone. We recommend you consider this module only if already have some experience in writing code or scripts, or if you have an appetite for learning.

Pre-Requisites

In addition to having an aptitude for analytical work and learning, the following pre-requisites apply to this module:

  • Completion of Modules 1 and 2
  • Completion of at least one Modelling Project of your own with Juno Cassandra under our mentorship (Module 3)
  • You have followed all lectures in the YouTube series C# for Civil Engineers, or alternatively completed another online course related to C# and Visual Studio.

Topics Covered

Part A - Building a Toy Model from Scratch

  • Overview of the Cassandra C# Interface for domain models
  • Overview of the Toy Model (input data, parameters, treatments, lookups)
  • Coding the Initialisation module
  • Building your project and running it in Cassandra
  • Coding the Increment module
  • Coding the Trigger module
  • Coding the Reset module
  • Coding the Routine Maintenance module
  • Debugging your model
  • Next steps

Part B - Extending the Default Road Network Model

  • Downloading the open-source Default Road Network Model
  • Reb-building the DLL file and linking it to Cassandra
  • Initialisation-class walk-through
  • Increment-class walk-through
  • Candidate selection class walk-through
  • Treatments-trigger-class walk-through
  • Reset-class walk-through
  • Routine maintenance trigger class walk-through
  • Making modifications and re-building your domain model
  • Debugging your model in Visual Studio while Cassandra runs
  • Best practices and advanced topics

Skills Gained

After finishing this module, you should be able to:

  • Have in-depth understanding of how Domain models work and how they link to Cassandra at run-time
  • Maintain and modify an existing domain model
  • Start coding your own Domain model from scratch in Visual Studio