Analytics has transformed retail experiences, mapped pathways for trains and trucks, allowed for the discovery of extraterrestrial life, and even predicted diseases. However, the inroads made in individual and social betterment are sometimes derailed by human error. While certainly not a new problem in analytics, their impact is no less devastating, causing the destruction of assets and financial losses in the billions of dollars in recent years alone.
When we boil it down, errors in reading, processing, analysing and interpreting data are fuelled by factors such as lack of experience, fatigue or loss of focus, lack of knowledge, or unacknowledged biases.
Artificial intelligence (AI) can help organisations overcome these by taking the front seat in parsing, analysing, drilling down, and dissecting impossibly large volumes of data. AI can also perform high-level arithmetic, logical, and statistical functions at a scale that would otherwise be impossible through human-led, self-service analytics alone.
Below are five of the most common human errors that can be eliminated using AI.
1. Confirmation bias and data manipulation
While bordering on cliche, it is no less true that rose-tinted glasses are a distinctly human trait that hinders objectivity. Confirmation bias narrows out searching, interpreting, and recall abilities.
See also: Conducting secure data movements in the cloud symphony
In the business world, it is easier for organisations to trust their gut instincts. They even go as far as to manipulate, omit, misrepresent, misinterpret the data to concur with their beliefs. On the other hand, artificial intelligence can use historical data to look for trends, patterns, and outliers, providing accurate, bias-free results.
AI-driven analytics provides automated insights into IT operations within an organisation. This helps save time on decision-making over IT operations – such as correlating incidents, planning and tracking projects, streamlining IT processes, and distributing staff to where they are most needed.
2. Inability to organise and use data effectively
See also: 80% of AI projects are projected to fail. Here's how it doesn't have to be this way
Data-related issues like organising multiple data sources, a lack of collaboration between data sources, low data accuracy, and poor data accessibility make it harder to derive true value out of data.
AI-powered solutions can resolve this issue by communicating with and correlating large data sets from several applications, databases, or data sources using relational data modelling techniques.
With nearly one million vehicles on the road, preventing traffic congestion is a critical objective for Singapore. Harnessing technology to do this, the Land Transport Authority (LTA) partnered with the Agency for Science, Technology and Research (A*STAR) to develop an AI-driven smart traffic light control system.
Dubbed the Cooperative and Unified Smart Traffic System (CRUISE), it uses a network of sensors that harness real-time datasets from global navigation satellite systems and autonomous vehicles. The data is then utilised in adjusting traffic light and pedestrian crossing timings for smoother traffic and passenger flow.
3. Downplaying losses
It is human nature to be loss-averse. One example is a 2017 incident involving two MRT trains crashing into one another at the Joo Koon MRT station, which injured around 38 people. The statement, made by the SMRT Corporation, claimed one of the trains "moved unexpectedly" and "came into contact" with another train. The company was severely criticised for trying to downplay the incident.
Earlier in 2012, the Public Utilities Board was also criticized for making a similar statement downplaying the 2012 flooding at Orchard Road.
To stay ahead of the latest tech trends, click here for DigitalEdge Section
The bottom line is that downplaying loss can be a costly mistake as organisations will suffer a loss of trust and will not be able to make effective decisions. By interpreting the data as it is, AI-driven analytics eliminates the tendency to favour positive outcomes over negative ones. This makes AI an ideal ally for leaders looking to make decisions based on complete facts rather than a partial picture.
4. Poor forecasting
Poor forecasting of the future is also another issue in human-driven analytics. Inflating budget requirements or underestimating supply and demand can cause businesses to lose profits and cancel orders. On the contrary, AI-led analytics tends to be more accurate because it makes predictions based on driving or arresting forces and external or environmental stimuli.
In healthcare, the use of AI can help doctors to predict disease progression. Changi General Hospital recently developed an AI predictive tool to calculate the risk of pneumonia patients requiring urgent care.
Developed using 3,000 chest and lung X-ray images and 200,000 data points, the Community Acquired Pneumonia and COVID-19 Artificial Intelligence Predictive Engine (CAPE) provides early warnings to doctors before a pneumonia patient becomes critically ill. This way, doctors can prescribe appropriate treatments to their patients on time.
5. Mired in surface-level analytics
Drilling down to analyse the root cause of problems can give businesses a headstart over their competitors. Root cause analysis can identify the cause of the problem, hint at remedial measures, and offer preventative measures that keep the problem from occurring in the future.
But with too many data sources, structures, and silos, it becomes impossible for humans to collate, analyse, and drill down to perform root cause analysis. AI-driven analytics can bypass these barriers by effortlessly digging into multiple levels of data simultaneously. Additionally, AI can also overlay several possible scenarios to come up with the most probable cause of a problem.
Ride-sharing service Grab and the National University of Singapore (NUS) launched an AI laboratory to develop solutions that will transform cities and public transportation in Southeast Asia.
By leveraging vast amounts of data collected by Grab, the laboratory aims to focus on improving transportation efficiency and creating algorithms that will better serve its customers. Furthermore, the laboratory will also expand to solve larger, world-impacting challenges such as traffic congestion and liveability in cities.
The future of analytics is AI
From providing actionable insights to eliminating errors or biases, AI-driven analytics allows organisations to stay ahead of the competition. Now that a growing number of business leaders are turning to AI to get insights that propel their businesses, we can expect AI in analytics to proliferate in Asia Pacific and globally.
Sailakshmi Baskaran is a product consultant at ManageEngine
