Simple tools for complex systems

Simple tools for complex systems

Simple tools for complex systems

J.P. Morgan & Chase

A tool for financial analysts to simulate the impact of essential service outages.

Role

Reporting & Analytics Lead

Timeline

1 Month

Team

Bill Hess (Design Lead) Meghan Hensley (Disaster Management Specialist) Eric Albright (PM)

Role

Reporting and Analytics Lead

Timeline

8 months

Team

Bill Hess (Design Lead) Meghan Hensley (Disaster Management Specialist) Eric Albright (PM)

Role

Reporting & Analytics Lead

Timeline

1 Month

Team

Bill Hess (Design Lead) Meghan Hensley (Disaster Management Specialist) Eric Albright (PM)

The J.P. Morgan & Chase Asset Outage Modeling Tool, or AOMT, assists analysts with emergency and disaster preparation planning and compliance. It is used by disaster management teams to model outage scenarios and identify potential cascading failure modes for critical financial infrastructure.

AOMT automates common risk analysis workflows and calculations, freeing up valuable mindshare for analysts to focus on creating precise and thorough strategic plans.

I worked end-to-end, from strategy, to scoping and design, to final delivery, guiding the reporting & analytics work stream and communicating with the client.

Research

The early stages of this project focused on extensive domain and user research to understand the disaster management landscape, including the specific challenges and requirements of J.P. Morgan's own disaster management team.

This involved user interviews, workflow observation sessions, and studying government regulations for disaster readiness. Below, you can see some my research untangling the complexity of how dependencies work.

Scoping the work

Understanding existing workflows, regulatory requirements, and individual users was crucial for scoping the design work. Through my research, I was able to infer additional feature requirements and optimize the layout and interactions based on the needs of real analysts.

I worked closely with the product owner to gather and understand business requirements and with a disaster management specialist to prioritize features.

Navigation

To navigate easily between criteria selection, applying filters, and viewing dependencies, the panel layout needed to be flexible.

Custom Inputs

To account for all the different asset types, as well as the specific assets themselves, the criteria panel needed a custom input for asset selection.

Panel Access

To keep track of selected criteria and edit criteria even after generating a model, the criteria panel needed to be accessible without leaving the tool.

Affected Items Display

To triage the most impacted items, the dependency viewer needed to display directly affected assets, with the option to display dependencies.

Bi-directional Linkages

To check and analyze dependencies, the dependency viewer needed a visualization of bi-directional linkages that could be understood at a glance.

Drill-ins & Downloads

To continue analysis in Excel, the dependency viewer needed to allow for drill-ins and data table downloads at any point while using the tool.

Panel Access

To keep track of selected criteria and edit criteria even after generating a model, the criteria panel needed to be accessible without leaving the tool.

Affected Items Display

To triage the most impacted items, the dependency viewer needed to display directly affected assets, with the option to display dependencies.

Workflow

This workflow centers on two tasks:

  1. Performing a search for various assets or compliance criteria, to view an overview of the model.

  2. Mapping the interdependencies of the resulting data, to understand potential cascading failures.

Example: Modeling disaster plans for a hurricane

To remain focused on a real use case while designing the interface, I focused on a hypothetical scenario:

As a disaster management analyst, I've been tasked with modeling potential disruptions to Florida's critical infrastructure in the 2024 hurricane season, so I can prepare a strategic overview of our preparedness plans.

As a disaster management analyst, I've been tasked with modeling potential disruptions to Florida's critical infrastructure in the 2024 hurricane season, so I can prepare a strategic overview of our preparedness plans.

As a disaster management analyst, I've been tasked with modeling potential disruptions to Florida's critical infrastructure in the 2024 hurricane season, so I can prepare a strategic overview of our preparedness plans.

As a disaster management analyst, I've been tasked with modeling potential disruptions to Florida's critical infrastructure in the 2024 hurricane season, so I can prepare a strategic overview of our preparedness plans.

Working against this scenario made it easier to generate accurate notional data to present to the client, and helped me understand outage modeling enough to create the scope for design and implementation.

Add Criteria: Criteria Type - State and Asset ID - Florida

Click Create Model

Review Model: Outage for State of Florida

Inspect ranked direct impacts and supporting assets

Click on number to open modal with downloadable table of information

Download data table(s) and close modal

Select row in table to begin inspecting dependencies

Use Dependencies Selector to view items that selection is related to

Review generated dependencies table(s)

Open downloaded data tables in Excel to continue analysis

Add Criteria: Criteria Type - State and Asset ID - Florida

Click Create Model

Review Model: Outage for State of Florida

Inspect ranked direct impacts and supporting assets

Click on number to open modal with downloadable table of information

Download data table(s) and close modal

Select row in table to begin inspecting dependencies

Use Dependencies Selector to view items that selection is related to

Review generated dependencies table(s)

Open downloaded data tables in Excel to continue analysis

Add Criteria: Criteria Type - State and Asset ID - Florida

Click Create Model

Review Model: Outage for State of Florida

Inspect ranked direct impacts and supporting assets

Click on number to open modal with downloadable table of information

Download data table(s) and close modal

Select row in table to begin inspecting dependencies

Use Dependencies Selector to view items that selection is related to

Review generated dependencies table(s)

Open downloaded data tables in Excel to continue analysis

Analysts need to modify their models without starting from scratch, viewing in real-time the impact their criteria modification would cause. To enable this workflow, I pushed for a two-pane layout allowing analysts to reactively refine search queries and filters by interacting with the dependency pane.

Core workflows

1

Performing searches with the criteria panel

The criteria panel allows analysts to iteratively query for general types of assets, as well as specific assets themselves. Because users needed to constantly refine their criteria, the panel is always visible on the left-hand side, with tabs to quickly flip between selecting criteria, applying filters, and viewing dependencies.

The criteria panel allows analysts to iteratively query for general types of assets, as well as specific assets themselves. Because users needed to constantly refine their criteria, the panel is always visible on the left-hand side, with tabs to quickly flip between selecting criteria, applying filters, and viewing dependencies.

2

Mapping impacts with the dependency selector

Based on the criteria selected, the dependency selector ranks affected assets by severity, displaying the chain reaction of potential disruptions in an interactive tree. This data is then exported to Excel for more comprehensive analysis and sharing with the broader organization.

3

Moving into Excel with data table downloads

To optimize analytical capabilities, most users would export data from the Asset Outage Modeling Tool to Excel. Given that many users are already familiar with Excel and utilize it in their current workflows, this integrated approach streamlines tool adoption while amplifying analytics capabilities and improving the user experience.

2

Mapping impacts with the dependency selector

Based on the criteria selected, the dependency selector ranks affected assets by severity, displaying the chain reaction of potential disruptions in an interactive tree. This data is then exported to Excel for more comprehensive analysis and sharing with the broader organization.

3

Moving into Excel with data table downloads

To optimize analytical capabilities, most users would export data from the Asset Outage Modeling Tool to Excel. Given that many users are already familiar with Excel and utilize it in their current workflows, this integrated approach streamlines tool adoption while amplifying analytics capabilities and improving the user experience.

3

Moving into Excel with data table downloads

To optimize analytical capabilities, most users would export data from the Asset Outage Modeling Tool to Excel. Given that many users are already familiar with Excel and utilize it in their current workflows, this integrated approach streamlines tool adoption while amplifying analytics capabilities and improving the user experience.

3

Moving into Excel with data table downloads

To optimize analytical capabilities, most users would export data from the Asset Outage Modeling Tool to Excel. Given that many users are already familiar with Excel and utilize it in their current workflows, this integrated approach streamlines tool adoption while amplifying analytics capabilities and improving the user experience.

Outcomes

I was able to unite stakeholders around a common understanding of user needs and UX best practices. Insights provided by my early research led to an accurately prioritized, cost-effective feature set that clearly addressed users' pain points.

The client's project manager left glowing feedback to that effect:

...I want to recognize Fiona for her work on the [JPMC ERMA] Asset Outage Modeler. She has enthusiastically accepted the challenge of creating a seemingly million wireframes for the modeler as the requirements are refined. Her wires were leveraged today for the presentation to the LOB/CF representatives with great success.

— Meghan Hensley (Project Manager)

...I want to recognize Fiona for her work on the [JPMC ERMA] Asset Outage Modeler. She has enthusiastically accepted the challenge of creating a seemingly million wireframes for the modeler as the requirements are refined. Her wires were leveraged today for the presentation to the LOB/CF representatives with great success.

— Meghan Hensley (Project Manager)

...I want to recognize Fiona for her work on the [JPMC ERMA] Asset Outage Modeler. She has enthusiastically accepted the challenge of creating a seemingly million wireframes for the modeler as the requirements are refined. Her wires were leveraged today for the presentation to the LOB/CF representatives with great success.

— Meghan Hensley (Project Manager)

My work on this project made the client confident in the consultancy's abilities, leading to a 6-month contract extension and budget to add three members to the project team. With a strong foundation for the future, the client has also been able to add new features and grow usage of the tool.