We don't know what we don't know. It's a common refrain that echoes through the uncertainties of supply chain management. Yet, what if we actively sought to address this ambiguity? What if we turned our gaze and ‘looked back’ to bring clarity and visibility to the realms of the unknown?
Conducting a ‘look back’ in supply chain planning is a powerful approach to uncover highly relevant insights from the analysis of historical data. This action involves scrutinizing what information was available in the past to understand how it could have shaped more informed decisions. This is much more than utilizing historical data to influence a forecast; innovations in look back technology leverage powerful AI (Artificial Intelligence) to review the data, make connections from seemingly disparate data, and identify the impact radius across the end-to-end supply chain.
The value lies in the ability to identify patterns, leading indicators, and areas for improvement. The goal? To contribute to supply chain resilience through continuous learning, leveraging AI and machine learning algorithms, and refining predictive models.
Let’s explore how this retrospective analysis, coupled with the prowess of AI aids both in understanding historical insights and contributes to shaping a more adaptive and resilient future in supply chain planning.
Just for you:
The Practical Role of Machine Learning and Artificial Intelligence in Supply Chain
Machine learning and AI are revolutionizing supply chains. Driven by the convergence of increased data availability, advances in computing technology, and global and interconnected supply chain networks, innovative algorithms are transforming the future of supply chain planning.
Use Case Example: Heatwave and Power Usage
In a real-world example at a power company, the team faced a predicament during a heatwave. Despite planning for increased power usage, they failed to predict the extent of the demand and supply could not meet the demand. A retrospective analysis of previous months, or a look back, revealed a crucial missing piece of the puzzle: humidity.
The models in place did not account for the impact of humidity on electricity consumption. People weren't just using air conditioning to combat the rising temperature; air conditioning units had to remove high humidity for comfort. This realization prompted a valuable learning opportunity – a chance to enhance the existing model and make it more accurate.
This process of looking back goes beyond revisiting historical events, to uncover the data that might have been overlooked – data that could have played a role in contributing to the event.
Why is this exercise important? By acknowledging and analyzing the events that transpired, it provides information we possessed but failed to use. It's about gaining visibility into what we didn't know we had – a concept that holds relevance in supply chain planning especially amid an era of increased risk and uncertainty.
The Role of AI in Supply Chain to Enhance Hindsight
The core idea is to compare decisions made with eventual outcomes. In the power company scenario, the planning team had to make decisions regarding power usage during the anticipated heatwave. The look back, facilitated by AI and machine learning in supply chain planning, revealed that incorporating humidity as a factor significantly improved the accuracy of predictions. The turn back, as it's also referred to, is a powerful tool enabled by AI. It allows businesses to evaluate past decisions in light of their outcomes. Machine learning algorithms play a pivotal role in this process, identifying which factors were most relevant and helping develop more effective plans through post hoc analysis.
From Hindsight to Foresight
As we embrace this approach of looking back, the key question emerges – what information did we uncover regarding cause and effect that was previously unknown? The hindsight gained from this process is not just for reflection; it's a tool to identify predictors that could have been used in the past and should be incorporated into future planning.
Just for you:
In this continuous learning cycle, machine learning is instrumental in refining and optimizing predictive models. In the power company case, the discovery of the correlation between humidity and electricity consumption led to a more nuanced understanding of scenarios. By recognizing the interaction between humidity and temperature, the company could make more accurate predictions and plan accordingly.
A Supply Use Case: Anticipating the Egg Shortage
Consider the most recent egg shortage - a situation that impacted shoppers as well as many manufacturers who use eggs as part of their production recipes. A real-world pasta company was significantly impacted, as was the industry, by numerous egg shortages that have occurred over the past few years, with multifaceted causes.
This situation underscores the criticality of identifying these leading indicators early on to accurately anticipate potential disruptions. As egg shortages have been impacting production and profits, this pasta company sought to find indicators of future shortages following the 2023 egg shortage. Upon employing look back analysis, it was found that an increase in material costs coupled with decreasing supply reliability, evidenced by declining fill rates and increased lead time days, served as leading indicators of a larger supply disruption.
Factors such as the observed increase in rising energy costs, egg prices, and shorted orders are leading indicators of a likely supply chain disruption. The ability to sense these indicators early allows for proactive measures, like entering agreements to secure egg supply and considering alternatives like using powdered eggs, where applicable. By leveraging historical insights and predictive analytics, planners can effectively navigate the challenges and ensure a more resilient supply chain.
How Supply Chain Planning Software Guides Your Look Back
Performing a look back in your supply chain, coupled with the power of AI and machine learning, opens avenues for continuous learning. It's a key step to acknowledge what we knew but didn't use, learning from it, and applying those lessons to strengthen the supply chain.
By analyzing previous successes and failures, supply chain teams can refine scenarios and strategies, ensuring that past lessons contribute to a more resilient future.
Advanced supply chain planning software helps companies retrace their steps to analyze past decisions and leverage historical data for more adaptive and resilient supply chains. John Galt Solutions’ Atlas Planning Platform streamlines the retrospective analysis while employing dynamic feedback processing to compare decisions taken with the eventual outcome, allowing businesses to assess the effectiveness of their past decisions. This dynamic feedback loop enables teams to gain insights into the correlation between their chosen scenarios and the actual outcomes. It’s an iterative process that enhances the machine learning algorithms to continually adapt and improve; refining predictive models and developing more accurate plans.
The look back approach is another example of how the synergy of human insights and AI capabilities propels supply chain planning into a realm of proactive resilience and informed decision-making. As you navigate an increasingly complex supply chain landscape, let's guide your look back and help you perform the retrospective exercise as a proactive step to build supply chain antifragility and drive continuous improvement.