Advanced AI Solutions for Asset-Intensive Industries

Keeping your Facilities in Shape with AI-Enabled Solutions.

METALS�& MINING

ENERGY

OIL�& GAS

PULP�& PAPER

ABOUT MODEC

ABOUT SHAPE

SHAPE’S SOLUTIONS

SUCCESS CASES

LIGHTHOUSE IN A BOX

VALUE ASSURANCE

GET IN TOUCH

About MODEC

1968

MODEC was

founded in Tokyo

1977

FPSO Castellon:

First FPSO ever built

1986

FPSO Kakap Natuna:

First EPCI FPSO Project

1998

FPSO MODEC Venture 1:

First Charter and O&M FPSO Project

2003

Brazil’s Office

BRAZIL’S HISTORY: 14 UNITS DELIVERED + 3 UNDER CONSTRUCTION

2003

FPSO Fluminense

2003

Rio de Janeiro Office (Corporate)

Macaé Office (O&M)

2007

FSO Cidade de Macaé MV15

2007

FPSO Cidade do Rio de Janeiro MV14

2009

FPSO Cidade de Niterói MV18

2010

FPSO Cidade de Santos MV20

2010

FPSO Cidade de Angra dos Reis MV22

Santos Office (O&M)

2013

FPSO OSX-3

2013

FPSO Cidade de São Paulo MV23

2014

FPSO Cidade�de Mangaratiba MV24

2015

FPSO Cidade de Itaguaí MV26

2016

FPSO Cidade de Caraguatatuba MV27

2017

FPSO Cid.de Campos dos Goytacazes MV29

2021

FPSO Carioca MV30

2022

FPSO Guanabara MV31

2023

FPSO Almirante Barroso MV32

2023

FPSO Anita Garibaldi MV33

2024

FPSO Bacallhau

MODEC’s History

2022

FPSO Miamte MV34

Global presence

Main Offices

FPSO in operation (16)

FPSO under construction (4)

FSO in operation (3)

USA

Houston

MEXICO

BRAZIL

Rio de Janeiro

JAPAN

Tokyo

Employes

+3,000

SENEGAL

COTÊ�D´IVOIRE

GHANA

SINGAPORE

VIETNAM

AUSTRALIA

NEW ZEALAND

Global

employees

+6,000

Total

portfolio

23 vessels

+ 280 years of operational & maintenance experience

36 FPSO

11 FSO

Buzios 5

Marlim 1

Bacalhau

6 TLP

1 MOPU

TOTAL

12

SOUTHEAST�ASIA

11

WEST�AFRICA

6

PACIFIC

6

GULF OF�MEXICO

1

NORTH SEA

EUROPE

17

BRAZIL

EPCI EXPERIENCE = 53 PROJECTS (~30% TO BRAZIL)

About Shape

Who is Shape?

History of MODEC’s Digital Transformation:

The Global Lighthouse Network

MODEC become a global lighthouse of the world economic forum

1^st

Oil upstream

company

1^st

Site in

Latam

1^st

Japanese

company

2015-2017

BUSINESS TRANSFORMATION

2017-2018

PILOT OF DIGITAL TRANSFORMATION

2019+

EXPANDING DIGITAL TRANSFORMATION

• Culture and mindset shift
• Operational Excellence

• Started from 1 FPSO
• 10 of the first use cases for Predictive Maintenance
• Models deployed to the cloud platform

• 9 FPSOs “operate” the AI models
• 500+ Developing further solutions
• 250+ bn lines of data in our cloud platform

Shape's mission is to create sustainable business value through the combination of data, technology, and real-world industry experience.

* major industries where Shape already provide solutions and support.

Shape is a spin-off of MODEC

IN PARTNERSHIP WITH:

MARKET*

• Oil & Gas
• Metals & Mining
• Pulp & Paper
• Power Generation

CUSTOMER PAINS

• Production Downtime
• Operational Costs
• Safety Risks
• Energy efficiency
• Sustainability

Minas Gerais

Pará

Rio Grande do Sul

Mato Grosso do Sul

Rondônia

Paraná

Singapore

Sergipe

Rio de Janeiro

BRASIL

Rio de Janeiro

São Paulo

Alagoas

Australia

Senegal

Japan

Mexico

Guyana

BRASIL

OIL & GAS

POWER GEN.

METALS & MINING

PULP & PAPER

SHAPE’S OFFICES

SHAPE’S PROJECTS

Global employees

~100

Global�presence

Shape

Revenue

Growth

Value proposition

Shape’s value proposition drives business impact by serving clients in three key pillars.

Integrate available data into a single source of truth

​

• Integrate
• Transform
• Export

INTEGRATED�DATA PLATFORM

Transform data into insights that address business needs

​

• KPI Tracking
• Data & Model integrity
• Asset health monitoring

APPLICATIONS�SUITE

Close the gap between digital solutions and front liners

​

• Modeling services
• Engineering support
• Process transformation
• Application training

VALUE �ASSURANCE

Lighthouse: Reliability Suite

DBMS:

Safety & Integrity

VET:

AI-Based Engineering

Sylph:

Sustainability Suite

Shape’s integrated solutions:

Fully dialogues with the Client's assets, by using Advanced Analytics.

Who is Shape?

AI-BASED ENGINEERING

VET

RELIABILITY

LIGHTHOUSE

Shape’s solutions

SAFETY & INTEGRITY

DBMS

SUSTAINABILITY

SYLPH

RELIABILITY�Lighthouse

Lighthouse:

value proposition

Maximize the performance of your facilities, ensuring current and future competitiveness.

IMPROVE RELIABILITY

INCREASE ASSET AVAILABILITY AND OVERALL PRODUCTION

IMPROVE RESOURCE

MANAGEMENT

REDUCE GHG EMISSIONS

MITIGATE HUMAN

RISK EXPOSURE

REDUCE MAINTENANCE & REPAIR COST

Lighthouse: potential impacts

Lighthouse track record:

* Based on average production in Brazil and Brent @$90

$160M

Additional Production

estimated/year

up to

$1,3M

Repair Cost Saving per plant/year

15%

Overall Downtime Reduction

estimated

Network effect:

The tool is continuously expanding. New models' creation and new failure modes!

8.500+

sensors tags connected�to AI models

370 million hours

of structured data analyzed

11 Plants

fully connected

800+

equipment models�deployed

Lighthouse:�problem statement

The solution support on asset-intensive industries surveillance to ensure asset health and safety.

SWISS CHEESE MODEL FOR RELIABILITY MANAGEMENT

Failure Causes

Failure prevention

Failure not

prevented

All downtimes in an asset (e.g., equipment failures and pre-failures) happen through a combination of multiple operational and processes variables at the same time;

The Lighthouse solution reduces the gaps in reliability management, improving equipment and process health.

Digital Solution�Platform for

Reliability

Lighthouse is a growing platform to provide digital solutions for the industry, primarily focused on reliability.

From detecting a possible failure to monitoring sensors, the platform allows the user to monitor the conditions of their assets in intuitive panels.

MAIN APPLICATIONS AVAILABLE:

DIAGNOSIS AND ACTIONS

Identify failure mode or key fault-related variables issuing recommendations.

ALARMS CONFIGURATION

Allows the user to manage and create alarms.

Lighthouse

ASSET HEALTH VIEW

The Home screen allows the integration of multiple plants into a single platform for quick identification of potential failure risks. �On a single screen, the management team can oversee and compare the health performance of multiple assets.

SYSTEM VIEW

The analysis follows with a view of systems and subsystems, where the operator checks in an illustrated way the systems that need more attention.

A powerful tool that enables decision-making based on data analysis.

A powerful tool that enables decision-making based on data analysis.

Lighthouse

EQUIPMENT VIEW

Next, the operator has access to vision by equipment, making maintenance prioritization decisions based on data.

The detailing screen by equipment aggregates all the information of the same, discoursing the main information for classification of the health of the equipment.

• Notes, Orders and Maintenance Plans; Oil Analysis Reports, Vibration, Thermography; Equipment Attributes.

Lighthouse

RISK MONITOR

The alarm response monitor displays the analysis graph of the risk model. The model is developed to identify possible failure trends and generates an alarm when the risk reaches a predefined limit.

CONTRIBUTION OF VARIABLES

The model identifies the contribution of variables to the risk of failure. Giving the user a quick diagnosis of where to act to prevent failure.

Success Cases

Lighthouse:�Go Live!

Go Live!

How it works

• SHAPE HUB
• Data Platform

• LIGHTHOUSE
• Application Center

Integrate

​

​

• PLANTS DATA
• Data Sources

​

​

Asset Hierarchies

Event Data Sources

CMMS / ERPs

Data Sheets

Transform

Export

• Maintenance KPI Tracking

• Prescriptive Insights

• Data & Model Integrity

• Asset Health Monitoring

• VET

• Time-to-failure Prediction

• Operation Simulation

• Anomaly Detection

• VALUE ASSURANCE

Modeling

Services

Engineering

Support

Process�Transformation

Application�Training

Historians

IIOT Platforms�- Data Lakes

• AI Models Library

SAFETY & INTEGRITY�DBMS

DBMS: value proposition

BASE CASE

ONLINE BOWTIE

GUIDED RISK ASSESSMENT PROCESS

SAP AND AMOS DATA INTEGRATED

INTEGRATED CASE TO OPERATE

DIGITAL BARRIER MANAGEMENT SYSTEM

Bowtie and SCE files

CMMS maintenance work orders information

Equipment, barriers and MAH classification

Reactive approach to risk assess SCE impairment event, and Case to Operate – paper-based system

Shape's Digital Barrier Management System

allows seamless and online monitoring of safety barriers

Digital Barrier Management System (DBMS) currently�use as data sources…

• CMMS (Computerized Maintenance Management System) data for equipment – SAP or AMOS.

…to generate and update critical information to stakeholders across assets.

• DBMS calculates the status of each barrier and displays that in live bow-tie diagrams;
• Operations team and relevant stakeholders are well-informed to make data-driven decisions to, for instance, prioritize maintenance;
• Stakeholders can act upon the most critical barriers.

DBMS

Digital Barrier Management System - DBMS provides a vision of the integrity status of Safety Critical Elements (SCEs) in operation and their relation to mapped threats, top events, Major Accident Hazards (MAHs), and consequences in Bowtie diagrams and help teams act in advance.

With DBMS, operation team can quickly identify sources of risks and make the required adjustments before an accident occurs.

DBMS Routine

Vessel Owner & Vessel Operator

Operator Team

• Offshore: FPSO
• Focal point/User: OIM

Operator Headquarter

• Europa Office: Portugal
• Focal point/User: Process Safety Engineer

Operator Support team

• Onshore: Local Office
• Focal point/User: Asset Management Coordinator

Vessel Owner

• Headquarter Office

Vessel Owner and Vessel Operator main areas involved in the project will have access according to permit level.

DBMS tool holds a huge flexibility capacity to reflect an asset safety process methodology and routine elected. Shape works in a collaborative environment with local Process Safety and Operation teams, and with MODEC’s Global Process Safety area.

DBMS�development roadmap

* We are continuously striving to identify the most effective enhancements for our tool, and we are receptive to new concepts and proposals for DBMS.

Demo DBMS

DEMO

SUSTAINABILITY�Sylph

Sylph: Problem Statement

~90% of these emissions come from Gas Turbine Generators, which means that we need to improve the FPSO’s energy efficiency to reduce it.

Possible�GHG reduction initiatives

Scope 1 emissions sources

(tCO₂e)

Stationary Combustion (80~85%)

Most of the flaring is the result of unplanned shutdowns, so will be the scope of the Predictive Maintenance solution.

Flaring

(10~15%)

Optimization of operational practices, including ensuring that all equipment run at their optimal energy efficiency point.

Operational optimization

Capital investments in new technologies, including installing new equipment or replacing existing ones with “green” alternatives.

Capital investments

Help operators to track and reduce scope 1 GHG emissions of industrial plants, striking a balance between production and sustainability by optimizing operational practices.

FPSO example:

WELLS

Production Header

Oil System

Gas Plant

Power Generation

SIMPLIFIED DIAGRAM

Recap: Problem Statement

To improve the energy efficiency of FPSOs, we propose to focus on GTGs and the Gas Plant¹, balancing with production requirements.

1 Responsible for ~70% of the energy consumption.

CRUDE OIL TANK

Recommend optimal pressures and flow rates for each compressor via adjustment of throttle and anti-surge valves, constrained by production, project specifications (e.g., Molsieve design pressure), and non-controllable variables (e.g., well pressure).

Recommend turning redundant Main A compressors on and off based on monitoring real vs. nominal flowrates.

Recommend repairs or switch to redundancies based on efficiency loss vs. theoretical performance curves provided by manufacturers.

The tool's first stage is centered on optimizing the Gas Plant. Optimization levers that can be considered by the Gas Plant VET

Sylph: potential impact

The initial estimation indicates a potential energy consumption reduction of at least ~5%.

Sylph: look & feel

Tool user-journey, starts with fleet-wide coverage and full-scale vessel optimization.

FLEET VIEW

A first visualization to support fleet surveillance

VESSEL VIEW

Expand vessel-level view with support for systems and their respective KPIs

Illustrative mockups

Sylph: look & feel

To contextualize each equipment virtual engineer module, we aim to include them in a vessel-level view.

THIS VESSEL-LEVEL VIEW CAN CONTAIN ELEMENTS SUCH AS:

KPI tracking: Emissions and energy efficiency-related metrics, in addition to important metrics for each covered system (e.g., GTG and ISO50001 KPIs).

Consolidated results: Calculated impact of implemented recommendations.

System drill-down: Access to system-level opportunities for all developed virtual engineer module.

Illustrative mockups

Sylph: look & feel

The virtual engineer module, for each system, aims to provide recommendations for different proposed optimization levers. We may recommend starting with most offensive systems.

Equipment virtual engineer module:

Illustrative mockups

SURVEILLANCE VIEW

System - and equipment - level metrics will allow users to understand their current state

OPTIMIZATION VIEW

Users will be able to generate ad-hoc recommendations or configure alarms, customizing production and operational constraints

Sylph: look & feel

To help the operations team track recommendation execution and impact, we propose to include a management module.

Illustrative mockups

The effect of each recommendation will be monitored and consolidated as the results of the tool.

Every recommendation issued by the solution will be tracked for whether they are accepted and executed as expected.

Sylph: capabilities summary

The first stage of this tool is centered on the optimization of a specific system.

Asset View

Equipment & Process View

Recommend optimal equipment setpoints – e.g., pressures and flowrates

Recommend turning equipment On / Off based on real & nominal parameters

Recommend repairs or switch to redundancies

based on efficiency loss vs. theoretical performance

Sylph has an estimation of expected impact, so that operations teams can prioritize and execute them

Expected impact could be used to estimate de ROI of investments

Fleet View

AI-BASED ENGINEERING�Virtual Engineering Tool - VET

SRU Virtual Engineer Tool

PROBLEM STATEMENT

The SRU membrane gradually clogs in operation and needs to be cleaned. Cleaning is expensive, takes time and consequently generates unavailability.

OBJECTIVE

Simulate the combination of CIP (Cleaning in Place) parameters to optimize the cleaning process, reduce the use of chemicals and provide longer campaigns, reducing downtime.

SOLUTION

A Virtual Engineering Tool that uses a CIP simulator to test various combinations of different parameter values ​​and recommend an optimal combination. This will enable the user to improve the performance of the CIP, resulting in important financial results.

SRU Virtual Engineer Tool

Using the operational knowledge, we can develop several virtual engineers trained specifically for the system.

OIL TREATMENT

MOLSIEVE

CO₂ REMOVAL

SRU

Recovery unit of sulfate

PdM models being improved to make process optimization recommendations.

The work begins with the objective of extending the duration of the campaign based on hypotheses raised by the operation.

What variables influence the dP of the membrane?

Temperature

Season

Flow

Opening PCV valve

Information from mediafilter (bacteria proliferation trend)

Sulfate

Sulfide

pH

Other process variables

Chlorine

SDI

Cartridge filter mesh size

Quantity of biocide (flow or dosage)

Frequency of biocide injection

Duration of biocide dose (ppm/hour)

Variables that influence bacteria proliferation

Campaign duration

Process variable

Lab data

CIP flow (alkaline phase)

Type of cleaning (normal or deep)

Chemicals used

Peaks of pressure during CIP

Process data from CIP auxiliary system

Duration of CIP (and the different CIP steps)

Process variables

Variables during the campaign

Variables during cleaning

What affects dP during cleaning?

What makes dP grow faster or slower?

What makes a high-performance CIP?

Problem statement: optimizing the CIP to reduce the total number of cleanings in a given period.

CONTROLLABLE VARIABLES

NON-CONTROLLED VARIABLES

INPUT

ACTUAL STATUS

USER INPUT

PREDICTIVE MODEL

VALUE ADDED

dP ESTIMATE MODEL THROUGH SELECTED PARAMETERS

FUTURE STATUS

dP SIMULATION

SOLUTION OUTPUT

We have developed a Virtual Engineering Tool to support the CIP planning decision process

Virtual Engineer Tool:

Simulador CIP

We have developed a Virtual Engineering Tool to support the CIP planning decision process

Virtual Engineer Tool:

Simulador CIP

DEMO

VET: look & feel

Success Cases

Use Case: Electric Motor

Electric motor internal component defect

EVENT DESCRIPTION

The model identified an early upward trend in the stator temperature without correlation with load increase in a HV electrical motor. After troubleshooting, a fallen internal air deflector plate was found.

OPERATION ACTION

The Equipment Specialist asked the offshore team to inspect the electric parameters and the cooling water condition at the electric motor`s heat exchange.

POTENTIAL SAVING

Risk scenario​: Water injection system partial stop, and in a worst-case, a mechanical damage in the fan or a short-circuit on the stator due to overheating caused by non-uniform air flow, with a total repair estimated cost of USD 600,000 in a worst-case scenario.

Use Case: Sea Water�Injection Pump

Seal failure on water injection pump

EVENT DESCRIPTION

Sea water injection pump with several historical events of seal failure and no clear root cause addressed before the PdM tool implementation. Just after the rollout for the MV23, several cavitation events were identified and associated as root cause of the seal problems.

OPERATION ACTION

The Equipment Specialist recommended an inspection of the thrust and journal bearing and the mechanical seal during the shutdown. Besides that, before, shutdown, adjust equipment setpoints (flowrates) and keep monitoring.

POTENTIAL SAVING

Risk scenario​: Water injection system partial stop, and in a worst-case, a mechanical damage in the fan or a short-circuit on the stator due to overheating caused by non-uniform air flow, with a total repair estimated cost of USD 600,000 in a worst-case scenario.

EVENT DESCRIPTION

The risk score started an upward tendency and triggered the alarms on PdM tool (figure below). Therefore, client specialist and Shape´s team analyzed this model and confirmed that the pressure on the oil side was out of limited area (figure above), which are the dark red points, indicating the start of scaling. The model allowed the team to identify the scaling occurrence at an early stage. This model already alarmed before as we can see on the risk score below and the specialist detected early signs of scaling. After the flushing, the risk started to increase again as explained above.

Use Case: Crude Heat Exchanger

OPERATION RECOMENDATION

As the flushing did not work as expected at the first time the PdM alarmed, this time the specialist recommendation was to exchange the plates.

POTENTIAL SAVING

Risk scenario​: Well shutdown for cleaning, resulting in 3 days of 100% downtime and with a total repair estimated cost of USD 1,694,000 in a worst-case scenario.

Scaling on Crude Crude Heat Exchanger

EVENT DESCRIPTION

VRU Gearbox 1^st stage bearing temperature transmitter (TI-2110-20) readings are presenting signs of malfunction. The temperature transmitter readings are below ambient temperature while the compressor is not operating, and during operation, the readings show a 35 °C difference from sensor TI-2110-21 that is placed on the same bearing.

Use Case:�VRU Compressor

OPERATION RECOMENDATION

Junction box, cables and fittings must be checked. In case of damage to parts, they should be replaced or fixed. Temperature transmitter (TI-2110-20) must be checked and calibrated. If calibration is not able to be properly performed, the instrument must be replaced.

POTENTIAL SAVING

Risk scenario​: Equipment sudden stop due to trip could lead to equipment internal parts damage (i.e., bearing seal, shaft). Potential downtime impact with a repair estimate cost of USD 350,000 in a worst-case scenario.

Gearbox 1^st Stage Temperature Transmitter Malfunction

Lighthouse in a Box

Accelerating the Time to Value with Out-of-Box Predictive Models.

Lighthouse in a box

Out-of-Box Predictive Models * On-Demand (150+ Models)

Equipment Surveillance Downtime avoidance

Reliability Platform with alarm Management

GTG

CENTRIFUGAL COMPRESSOR

CENTRIFUGAL PUMP

SCREW COMPRESSOR

MOTOR

FAILURE MODE

PERFORMANCE

FAILURE CAUSE

HEAT EXCHANGER

* AI-based models require retraining services.

Sample of Out-of-Box Predictive Models Matrix.

Lighthouse in a box

ALARM

Pre-defined thresholds per equipment

Shape and client work in partnership to customize application and develop new models

Beyond Lighthouse in a box, in a collaborative basis Shape and client work together to increase value added.

Complement the Solution deployed according to client`s particular needs and potential in-house initiatives

Specialized fit-for-purpose Model Creation

PaaS

SaaS

SaaS

Shape’s Model Library

150+ Equipment Models Online

SAMPLE OF EQUIPMENT TYPES

• Centrifugal Compressors
• Screw Compressors
• Centrifugal Pumps 
• Electric Motors
• Gas Turbines - GTG

Business Models:

Lighthouse in a box

HIGHLIGHTS

• Continuous Support and Maintenance
• Blueprint Archtecture
• Cloud Costs Reduction (TCO)

GENERAL ARCHITECTURE – SHAPE CLOUD:

Data lake at customer`s cloud and model processing data visualization and diagnosys proccess at Shape`s cloud.

Maintenance

Data Source

(ERP / CMMS)

Data Historian

Real-Time Data

Customer

Data Lake

Shape

Hub Lake

Application

App Data

Warehouse

CUSTOMER CLOUD

SHAPE CLOUD

Data

Processing

MLOps and�App Standards

Lighthouse in a box

How it works

• Shape Hub
• Data Platform

• Lighthouse
• Application Center

• Value �Assurance

• Apps

• Analytics

• VET

• Maintenance

• Process transformation

• Application training

• Maintenance KPI tracking

• Maintenance Insights

• Time-to-failure prediction

• Operation simulation

• Modeling services

• Engineering support

• Anomaly detection

• Data & model integrity

• Asset health monitoring

• Integrate

• Export

• Transform

Asset Hierarchies

Event Data Sources

CMMS / ERPs

Equipment Data Sheet

Historians

IIoT Platforms / �Data Lakes

TIME SERIES DATA

​

​

• Plants Data
• Data Sources

​

​

Value Assurance

Shape Value Assurance

INTEGRATED DATA PLATAFORM

DATABASE

*ETL + CONTEXTUALIZATION

CLOUD

Data Lake Processed data ready to use

APPLICATIONS SUITE

• Temperature

DATA

PREDICTIVE MODELS

• Autoencoder
• Classification
• Regression

Mathematical Equations

• Flow

• Vibration

RISK SCORE (interpretation)

• Time to Failure
• Failure probability
• Remaining useful life

ALARM

• Pre-defined thresholds per equipment

Engineers and Data Scientists

building models together

Shape provides engineering support + IT support since failure mode discover

VALUE ASSURANCE

SMS

E-Mail

Push

Notification

OPERATIONS ENGINEER

DIAGNOSTIC SCREEN IS CUSTOMIZABLE PER EQUIPMENT

Failure Mode Discovered

FEEDBACK

Execution Follow-up

• Prioritization
• Due date

Value Assurance

• Uptime increase
• Maintenance cost reduction
• CO² emission reduction

Shape Value Assurance

REMOTE CONTROL CENTER

AI-MODELS POWERED BY ENGINEERING

Data Science team working side by side with engineers building AI models leveraging industry knowledge.

USER SUPPORT FOR FPSO OPERATOR

Our engineers work side-by-side with clients’ teams, supporting them to better use the tools we provide.

OPPORTUNITY IDENTIFICATION

Shape’s team is constantly identifying opportunities for new solutions and improvements – models network effect within MODEC’s fleet.

We have Shape engineers working aside clients with AI-based technologies.

Shape Value Assurance

Shape’s

Performance

Report

Reports will be delivered, detailing occurrences and the financial impact of Lighthouse.

Report

Get in touch

If you have any questions, comments or just want to say hi, feel free to get in touch.

Simply scan one of the QR codes on your screen using your smartphone's camera.

WEBSITE

E-MAIL

LINKEDIN

Failure modes and causes study

Methodology:

Failure Modes�“symptoms”

Failure Causes�“Root Causes”

P-F Curve

Health Score

P-F Curve

Health Score

Time

Time