Why AI-optimized workflows aren’t always best for businesses

Workflow and process inefficiencies can add up to 40% of a company’s annual turnover. In many cases, companies try to solve this problem by implementing artificial intelligence (AI) scheduling algorithms. This is seen as a beneficial tool for business models that rely on speed and efficiency, such as delivery services and the logistics sector.

While AI has certainly helped with some of the time-consuming and often unpredictable tasks associated with scheduling employees across departments, the model is not yet perfect. Sometimes it makes problems worse and not better.

AI lacks the human ability to see beyond just optimizing for business efficiency. That means it has no capacity for “human” variables such as employee preferences. The limitations of AI scheduling can often lead to unbalanced shifts or dissatisfied employees, culminating in situations where the AI ​​”help” given to HR gets in the way of smooth workflows.

When optimization goes wrong: AI can’t see the human behind the data points

Auto-scheduling AI has gained tremendous popularity in recent years. Between 2022 and 2027, the global AI scheduling system market is expected to reach a CAGR of 13.5%and 77% of businesses are already using AI or want to add AI tools to optimize workflows and improve business processes.

However, it is important to note that AI cannot yet create schedules without human supervision. HR professionals still need to review and adjust auto-generated schedules because there is still a huge, glaring flaw in the AI ​​algorithms: a lack of “human parameters.”

AI is great at sifting through data and finding ways to maximize business process efficiency. Workflow optimization through algorithms using historical data is ideal for projecting things like order volume and required number of employees, based on information such as marketing promotions, weather patterns, time of day, hourly order estimates, and average customer wait times.

The problem stems from AI’s inability to account for “human parameters,” which it sees as a drop in efficiency rather than better business practice.

For example, if a company has observant Muslim employees, they need small breaks in their working days to observe prayer times. If a company employs new mothers, they may also need built-in times to express breast milk. These are things that are currently beyond AI’s ability to properly account for, as it cannot use empathy and human reasoning to see that these “inefficient schemes” are much more efficient from a long-term employee happiness perspective.

Efficiency is not always the best policy; is there a solution?

Currently, automatic scheduling tools can only pull data points from limited sources, such as timesheets and workflow history, to evenly distribute work hours in what they believe is the optimal way. AI scheduling tools need help understanding why it’s bad to have the same employee work the last shift one day and return for the first shift the next. They also cannot yet account for individual employee preferences or varied availability.

One possible solution to this problem is to keep adding parameters to the algorithms, but that brings its own problems. First, every time you introduce a new parameter, it decreases the chance that the algorithm will perform well. Second, algorithms only work as well as the data they are given. If AI tools are provided with incomplete, incorrect, or inaccurate data, scheduling can hinder workflow efficiency and create more work for managers or HR staff. Adding more filters or constraints to the algorithm won’t help it work better.

So what’s the solution? Unfortunately, until we discover ways to infuse AI with empathetic reasoning, humans will probably always have to lend a hand in employee scheduling.

Nevertheless, companies can work to create a more positive, synergistic relationship between AI planning tools and the people who use them.

For example, delivery companies can feed historical data into AI tools to increase the effectiveness of their initial planning outputs. This relieves some of the burden for HR and scheduling managers. In turn, the human planner now has an optimized base schedule to work from, reducing the time spent fitting workers into the required time slots.

AI may be perfectly efficient, but it still needs human help to make employees happy

Humanity is still working hard developing AI that exhibits “general intelligence,” which is a term applied to the intelligence observed in humans and animals. It combines problem solving with emotion and common sense, two things that have yet to be replicated in AI.

When you need to automate repetitive tasks or analyze massive amounts of data to find inefficiencies and better ways of working, AI almost always trumps humans. However, once you add nuance, emotion, or general intelligence, such as in task scheduling, people will still need to have the final say in balancing optimized workflows with employee satisfaction and long-term business growth.

Vitaly Alexandrov is a serial entrepreneur and founder and CEO of Food Rocketa US-based express grocery delivery service.