Automation in context

Automation definition

Automation involves applying technology (tools) and processes (techniques) to perform tasks with minimal human intervention. Automation uses various tools, software, machinery, and systems to streamline and optimise workflows, reduce manual labour, improve efficiency, and enhance accuracy. Automation technologies can be used in building fully or partially automated decision-making (ADM) solutions.

The NSW ombudsman defines ADM solutions as computer-based systems that are used by an organisation in an administrative decision-making process affecting the public community.^4,5,6 ADM solutions influence or drive human decision-making. They provide insights, guidance or initial assessments, recommend a decision, or make the final decision.

The portfolio of automation types is extensive.⁷ It covers diverse categories such as integration Platform as a Service (iPaaS), Business Process Automation (BPA), IT Process Automation, Test Automation, IT Infrastructure Automation, Low-Code Application Platforms (LCAP), AI-Powered Automation and others.

In this guide, we prioritise Robotic Process Automation (RPA) and Intelligent automation (IA). RPA and IA solutions enhance productivity by automating or augmenting a manual work process.⁸

RPA enables quick fixes and captures tasks easy to automate.
IA automates complex tasks that require the combination of human cognitive abilities with IA’s artificial intelligence capabilities.

Robotic process automation (RPA)

Robotic process automation (RPA) refers to deploying software agents to automate simple, manual, and repetitive tasks traditionally performed by humans.^9,10,11 RPA automates tasks such as data entry and report generation, while providing versatility, scalability, and seamless integration with existing IT systems. RPA enables businesses to streamline operations and allocate human resources to more complex and meaningful tasks.

Intelligent automation (IA)

Intelligent automation (IA) means adding AI technologies to RPA.^10,11,12 This addition results in the augmentation of RPA. IA technologies include, but are not limited to, conversational AI, machine learning (ML), natural language processing (NLP), and process and task mining. They let automation handle complex tasks and augment RPA with cognitive abilities assisting or driving decision making. Unlike RPA, which handles rule-based tasks, IA comprehends unstructured data, adapts to change, and achieves automation by leveraging machine learning and AI components. A contemporary example is an AI-powered tool that uses machine learning to assist software developers write code.

Value for business

The process automation performed by RPA and IA technologies has the potential to significantly boost the NSW Government’s operational efficiency. Automation increases accuracy and adaptability and supports improved decision-making. RPA and IA features can overlap. The features you include in your solutions will depend on the complexity of the automated process and your organisational needs.

Expand the headings below to review the features, value, business benefits and skills in RPA (basic) and IA (advanced) required in context.

Technology capabilities and business value

Feature	Use and value	Business benefit
Screen scraping (basic)	Used to extract data from a computer screen for use in another application. Useful when working with legacy systems lacking APIs. Valuable for extracting data from user interface elements.	Accuracy and consistency Operational efficiency.
Workflow/orchestration automation (basic)	Used to automate simple tasks and processes. Useful for routine, repetitive tasks with clear rules. Valuable for automating rule-based processes.	Cost and time savings Enhanced productivity.
Rule-based decision making (basic)	Used to automate decisions using predefined rules and logic. Useful for tasks with predefined decision rules. Valuable for applying conditional logic such as if-then.	Accuracy and consistency Enhanced productivity Cost and time savings.
Task scheduling (basic)	Used to automate the assignment and timing of tasks and optimise workflow. Useful for time-sensitive or batch processing tasks. Valuable for scheduling and executing tasks efficiently.	Operational efficiency.
Integration connectors/APIs (basic)	Used to automate data exchange between systems. Useful when exchanging data between systems. Valuable for connecting with external systems and databases.	Accuracy and consistency.
Exception handling (basic)	Used to automatically managing errors and anomalies in processes. Useful for ensuring robust and fault-tolerant automation. Valuable when managing errors and exceptions.	Improved reliability.
Natural language processing (NLP) (advanced)	Used to enable machines to understand and interpret human language. Useful for chatbots, document analysis, and more. Valuable for understanding and processing human language.	Improved community satisfaction.
Machine learning (ML) (advanced)	Used to enable systems to learn from data and improve over time. Useful for tasks involving data analysis, prediction, or classification. Valuable when learning from data and improving over time.	Informed decision-making.
Cognitive automation (advanced)	Used to mimic human thought processes to enhance decision-making and problem-solving. Useful for complex processes requiring contextual understanding. Valuable for simulating human-like decision-making.	Informed decision-making.
Computer vision (advanced)	Enables machines to interpret and analyse visual information from images or videos. Useful for image recognition and extracting data from images. Valuable for interpreting visual information.	Accuracy and consistency.
Predictive Analytics (Advanced)	Uses data and statistical algorithms to forecast future outcomes and trends. Useful for decision-making in dynamic and data-driven scenarios. Valuable for analysing historical data to make predictions.	Informed decision-making
Speech Recognition (Advanced)	Enables machines with the ability to process spoken language. Useful in voice-activated systems or transcription tasks. Valuable for converting spoken language into text.	Improved service delivery Increased inclusive services
Sentiment analysis (advanced)	Analyses text to determine the emotion, tone and opinions expressed. Useful for social media monitoring, customer feedback analysis. Valuable for determining sentiment in text data.	Improved service delivery Improved community satisfaction.
Optical character recognition (OCR) (advanced)	Used to convert printed or handwritten text into machine-readable data. Useful for document processing and data extraction from scanned documents. Valuable for extracting and analysing text from images.	Enhanced productivity Cost and time savings.
Intelligent document processing (IDP)	Extracts data from documents. Useful for extracting, interpreting, and processing information from unstructured documents. Valuable when automating document processing workflows.	Operational efficiency Cost and time savings Accuracy and consistency.
Task and process intelligence (advanced)	Analyses workflows and tasks performance. Useful for analysing and optimising end-to-end process Valuable for analysing, modelling and optimising business tasks and processes.	Operational efficiency Improved decision making.
Generative AI (advanced)	Generates new content from existing content. Useful for generating new content learned from existing data. Valuable for generating software code, testing scenarios, data augmentation and personalised content creation.	Enhance productivity Cost and time savings Improved decision making.

Automation in context

Automation definition

Robotic process automation (RPA)

Intelligent automation (IA)

Value for business

Explore the NSW Automation Guide