What is AI?
Artificial intelligence (AI) is the simulation of human intelligence processes by machines, especially computer systems. The term AI is often used as an umbrella term for machines that are capable of perception, logic, reasoning, and learning.
Another useful way to think about AI is in terms of a spectrum from automation (rules-based) to intelligence (learning systems) applied to various problems or use cases.
Machines are good at repeating processes producing predictable outcomes. What machine learning provides is identifying patterns and anomalies; and then scale, repeat and produce consistent results. AI enables optimisation, intelligent automation, and augmenting human ability, or build systems that perform functions beyond the limits of human ability.
Our Vision
Mainstreaming AI: Translating AI into practice for companies.
Why AI?
- Deriving more value from your data
- Better business decisions
- Enabling new business models, products and services
- Reduction in risk and cost
- Increase in sales
- Discovering insights (“To make the unknown known”)
- Improvement of customer and employee experience
Our capabilities
We have a strong capability in AI and apply these capabilities to mainstreaming AI, strategy and solution development and integration. Our capabilities include:
- AI, Data Science, Advanced Analytics, Automation Robotics
- Pattern recognition, anomaly detection, predictive analytics, object classification
- Machine learning, deep learning, neural networks
- Broad & deep understanding of techniques and systems in the AI & AA space
- Data architectures, data management, data engineering, visualisation
- Systems architectures
- Road-mapping and strategy development
- Following an agile approach and use appropriate methodologies such as Design Thinking
Our Offerings
- AI Readiness Assessment, AI and Data strategy
- Robotic Process Automation (RPA)
- Knowledge Base for workforce automation and front-office process compliance
- Text-to-Speech in South African Languages
- Chatbots and Virtual Assistants
- Customer Specific AI and Advanced Analytics Solution Development.
Examples of AI solutions that can be implemented include:
- Customer targeting (Behaviour, risk profile, etc.)
- Process optimization (productivity, accuracy, speed)
- Trends, predictions and recommendations (churn, fraud)
- Intelligent Product Quality Assessment
Benefits of AI/Active in the following fields
Intelligent process automation (IPA)
By combining process automation, machine learning and natural language processing to routine activities it mimics, and augments tasks performed by humans and over time it learns to do them better.
This drastically improves efficiency and performance, reduce operational risk and frees time of humans to focus on higher value activities.
Inventory and parts optimisation
User demand, supplier orders, and inventory levels are all improved using machine learning and deep learning systems. Demand is more accurately predicted while reinforcement learning can act on these predictions.
This improves inventory teams’ efficiency resulting in cost savings in terms of unnecessary stock to be stored and reduced waste due to stock expiration.
Network management and optimisation
Mobile networks are becoming ever more complex, subscriber demands are increasing (reliability, coverage, bandwidth, security, etc…) and immediate action is often required.
Deep reinforcement learning AI on a combination of traffic, network and customer data, achieves a greater balance between network economics, performance, and end-user experience.
Predictive/preventative maintenance
When machinery in a production environment fails it results in revenue loss and a reduction in productivity.
Machine learning predicts the failure of machinery in a manufacturing process and/or the final product before it happens. This results in increased productivity reduced downtime and increased after-sales services.
Quality control
Quality and yield are major drivers in manufacturing.
Machine learning identifies and predicts defects and contributing factors across all processing stages of the factory. The identification of features to address is used to improve the overall quality, reduce waste, and optimize yield.
Single identity management
Organisations need to be able to uniquely identify the same entity (e.g. a person) over multiple internal and external systems to create a single view of a customer.
Machine learning augments the identification of a single entity (e.g. a person) over multiple, heterogeneous information sources and systems. This is applied towards improved service delivery, increased sales and personalisation.