The increasing skills shortage and the high cost of employing new talent are key challenges for Australian businesses.

The labour market’s unpredictability makes it vital for organisations to take advantage of their existing assets to maximise efficiency and minimise the risk of losing corporate knowledge if key, long-term employees quit.

An organisation’s business rules are a major asset, but their management is often undervalued, making many simple tasks more time consuming and complex than is necessary.

Most organisations store business rules in myriad locations. Many exist largely in the heads of the people who act on them. That means the rules are interpreted according to the recollections of those who work with them, which can lead to multiple variations.

Another source is rules that are hard coded into disparate technology systems, including customer relationship management systems and databases. Similar rules may exist in more than one place, but not all variants are updated when business processes or legislation changes.

The result can be a mishmash of rules data across an organisation with no central repository and therefore no streamlined, consistent methodology about how the rules are accessed and applied.

For example, an insurance company has defined claims management processes and limits, but the nuances of an individual claim’s treatment often depend on the interpretation and implementation of those rules by an individual claims assessor, who may seek assistance from a more experienced assessor.

The combination of hard-coded rules and those embedded in people’s heads makes it challenging and overcomplex for anyone attempting to understand what rules have been applied, where and how.

The business process and accompanying decision making is guided by individual employees based on their own experiences, combined with hard-coded values that may no longer match current legislation and other business requirements.

That creates several problems:

• Inconsistent customer experiences (employees applying rules differently)
• Lack of scalability (if queries rely on responses from the most experienced employee, that creates a bottleneck)
• Sluggish updating of business rules (due to a lack of employee training or the inability to quickly update legacy systems)
• Duplication of manual input (repetitive, single items of work requiring significant individual effort)
• If an employee leaves, it’s hard to retain their individualised decision making.

The solution is to capture the corporate knowledge and business rules and store and maintain them within a centralised repository, a low-code expert system or rules engine that integrates with your existing systems.

A few definitions:

Low-code – this employs visual interfaces with basic logic and drag-and-drop capabilities instead of complex programming languages. That means users don’t require deep technical expertise to build and update rules. Changes can be made by any computer-literate business employee.

Expert system – this is the rules engine that stores, centralises and applies all the business rules. It becomes the source of truth and ensures rules are applied consistently.

The advantages of a low-code expert system

• All channels apply the rules in exactly the same way. There is no different interpretation or experience depending whether you access the system via a website, mobile app or phone. Every channel uses the same core logic contained within the expert system.

• Rules are removed from hard-coded technology products and placed in the centralised repository, obviating the challenge of updating code in various locations in response to legislative or policy changes.

• Scalability – the expert system manages a large share of the decision making, regardless of the quantity of decisions required.

• The manual effort of decision making is reduced. People maintain control, but AI manages the bulk of the mundane simple decisions, leaving people to do what they do best – talk with customers, listen to what they need and help them understand the organisation’s decisions.

• Auditability – there’s no need to ask someone why they made a decision for audit purposes. The expert system shows exactly what rules were applied and how, leaving a clearly discernible pattern to show why a decision was reached.

RulesLab is data and analytics specialist PBT Group’s solution that standardises rules that reside within people’s heads and removes them from hard-coded technical products, managing them all together, within a low-code expert system.

Using AI, RulesLab frees skilled employees from labour-intensive, repetitive tasks, enabling greater speed, constancy and quality in the outcomes of decision making.

RulesLab is a decision-support mechanism, not black-box end-to-end automation. Humans always remain in control.

Organisations generally start small, with a series of rules, then build their rules engine in incremental steps, testing as they go.

RulesLab adds a layer of protection and removes the cognitive burden on team members having to remember myriad pieces of information and how those knowledge hubs interact with other pieces of information within the business rules.

Organisations control who can change rules or add new ones. Rules engines are applicable to any regular activities performed more frequently than weekly.

Three simple steps are involved in automating regular, routine decision making:

1. Capture the business rules.
2. Store and maintain them in a central repository.
3. Automate them.

Implementing RulesLab provides a central source of truth. It enables artificial intelligence (AI) to do what it does best and people to focus on high-level human interaction and customer support, without wasting time on routine, regular tasks that are easily automated.

RulesLab in health and aged care

Clinicians and aged care workers are frequently required to manage clients’ medication regimes.

As a client information database becomes more complex, more factors come into play about the potential reactions of medication combinations. There’s an expectation that clinicians know the ramifications of mixing different medications, but all humans are fallible.

RulesLab uses AI to ensure dosages are correct and patients’ individual circumstances considered. For example, a clinician forgetting to ask a simple question that elicits basic information, for example, whether a patient has diabetes, could mean the wrong drugs are prescribed with potentially dangerous ramifications.

When patients are assessed against a standard set of rules, ambiguity and inconsistency are removed.

RulesLab can assist in triaging patients in health and aged care, ensuring consistency according to the actual level of care required, rather than subjective factors.

In aged care facilities, agency staff are frequently unfamiliar with patients’ regular medication regimes. RulesLab reduces the opportunity for error.

Once rules are input to RulesLab, they are tested before real-world application and, in a clinical environment, overseen by a qualified medical practitioner before implementation.

RulesLab can assist health insurers to identify potentially fraudulent claims. For example, doctors overprescribing medications or dentists claiming for recurrent tooth extractions or too frequent x-rays.

Analytics across the data can intercept potential issues before they develop into serious problems.

RulesLab in insurance

A low-code rules engine can assist insurers to triage claims during catastrophic events and speed the claims management process.

RulesLab can identify simple claims that can be processed quickly with relatively low human intervention, and flag those that are more complex, likely to be more costly and require greater individual attention to detail. 

RulesLab can identify potentially fraudulent claims, avoiding the nightmare of trying to recover funds after a fraudulent claim has been paid.

Read more about RulesLab’s authoring tool here.

Read more about the components of RulesLab here.

Read FAQs about RulesLab here.