How the Theory of Constraints Applies to Robotic Process Automation

Or how automation without a plan and approach is a quick trip to chaos

Ian Barkin and David Brain, Symphony Ventures

In search of something new to say about the oncoming freight train that is Digital Labor, the team at Symphony, a professional services firm, has gone back to the fundamentals. And, in doing so, we’ve rediscovered the underpinnings of this new labor revolution that have always existed as part of the way we study complex systems in order to improve upon them. These underpinnings are constraints. In other words, automation is a new (and better) way to address enterprise constraints.

Our inspiration was a book we read nearly 15 years ago – The Goal by Eli Goldratt. For those not familiar with the premise, the book is about new principles of manufacturing in which Goldratt introduces the “Theory of Constraints” as a way to manage complex systems. Written in 1984, The Goal became a widely read business tome. So, why is it relevant for Shared Services Centers (SSC) and customers of Business Process Outsourcing (BPO)? A few reasons. First, its subtitle, – A Process of Ongoing Improvement – succinctly sums up what the entire SSC industry has branded itself around for the last decade. Second, the book and the approach was a mission for Goldratt to “change manufacturing from an art to a science.” This, we believe, is the role that Robotic Process Automation (RPA) will play in shared services and outsourcing. Finally, the ultimate lesson is about avoiding future constraints caused by inertia. This one will most certainly rear its head as a concept in the next 18-24 months and is worth exploring now.

A Process of Ongoing Improvement

As we’ve articulated in a Symphony blog titled “BPO = Business People Outsourcing,” the BPO (and SSC) industry has had a great run at improvements and evolutions from simple arbitrage to continuous improvement to analytics and insight, and so on.  At every stage, BPO and SSC providers have evolved to address the next pressing challenge. At first, it was high labor costs (arbitrage). Next it was inefficient processes (continuous improvement). Then it was a lack of a holistic understanding of a system and the resulting strategic blinders in place (analytics and insight).

In his novel, Goldratt uses his Theory of Constraints to highlight that the success of an entire factory (or any process at all) is determined by the choke point that constrains the overall output. In other words, the whole factory can only move as fast as its slowest resource. This bottleneck can occur at the beginning, middle, or end of a process. It could be an outdated machine, an inefficient worker, or a poorly designed company-wide policy. Goldratt suggests that, whatever the bottleneck is, every action should be taken with it in mind, and every resource should be directed at eliminating it. And this process should never stop. Once the worst bottleneck is identified and eliminated, the next culprit of inefficiency must be identified.

The general fervor around RPA in SSC would suggest that the next bottleneck in everyone’s crosshairs is ‘people.’ However, we believe that the approach of Digital Labor as human replacement is too simple, and frankly not very creative. We are more inspired by the role RPA can play in alleviating bottlenecks of Process and Technology. Process bottlenecks exist in the form of poorly built workflow, undefined business logic or simply heterogeneity of process resulting from acquisition, rapid growth and lack of process design mindset. Technology bottlenecks exist in the form of a spaghetti of legacy and homegrown systems struggling to work together to achieve business needs. In the process and technology bottleneck scenarios, an enterprise that thoughtfully deploys RPA is given a chance to not only emulate human tasks, but alleviate historical complexity inherited over time.

Art to Science

Here we share a particular kindred spirit with Goldratt. For too long, the design of workforce solutions has been a creative and flexible art – productivity levels, output volumes, retention rates, over-hire percentage, stabilization runways, knowledge capture methodologies, and more. The list of solution design components is long, and the rigor and precedent by which they are sculpted is loose. The last decade will be seen as the ‘art of (services) work’ decade. We dare say that Digital Labor has the ability and the momentum to change SSC and the outsourcing industry from one of art to one of science.

As we’ve seen, when we design a solution for a digital workforce (robots), we can no longer afford to be vague. It is incumbent on the solution design team to model, map and capture each end-to-end process to the most minute detail. This perhaps is where we are most fanatical. In working with our clients we must see, touch, record, and model every ounce of a process in order to clearly articulate judgment versus rote tasks, understanding the interplay between each branch of business logic in a process. It’s not good enough to do this for a discrete ‘automatable’ task. If an enterprise is not looking end-to-end, it’s laying the groundwork for convolutions and complexities in the near future.

Beware the next system constraint

The final warning in Goldratt’s Theory of Constraints is to not allow inertia to cause a system constraint. In other words, don’t let the work done previously coast on unchallenged so that it becomes a future choke point. This is particularly interesting, and why we believe orchestration in this new Digital Labor environment is as (or more) crucial than the careful design of the first robots in a system. The RPA-based ‘work’ industry (in both BPO and Shared Services) is unquestionably in its infancy. There are pilots, proofs of concept and in-house developments creating a maelstrom of press, promotion and promise. Without much doubt we can assume that the big BPO and Shared Services operations are test-driving numerous vendors. And, just as predictably, some of the big BPO houses are pretending to be software development shops, crafting their own robot code. While this exploration is healthy, the result of all this will be a robot Babel scenario. Choose any taxonomy – processs silos, business units, service contracts, delivery centers – the  story will be the same. There will be a lot of robots. Robots that will not be able to talk with one another. And unless good orchestration is in place from the beginning, the inertia of ‘automate everything’ will become our next bottleneck. And, perhaps fodder for our next hype cycle.

Eli Goldratt passionately pursued the conversion of ‘art to science’ by addressing constraints in order to unleash business potential. We believe, and have seen, how RPA can be a catalyst for this same ‘art to science’ shift across the enterprise work environment. However, in our work advising clients on their Digital Labor strategies, we have found that far too often, the rigor needed to succeed is not put in place at the very beginning of the journey. By following a modern approach, working with expert advisors and by requiring solution design to be specific and accurate, automation is enabling the adoption of the ‘Future of Work’ and driving a new level of process focus that will color a generation of task design and execution.

Source: How the Theory of Constraints Applies to Robotic Process Automation By: David Brain and Ian Barkin, Symphony Ventures

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s