Creating Supply Chain Resilience Through Accurate Demand Planning and Execution

When brands were asked to name their top supply chain investment strategies in the 2021 RIS "Supply Chain Technology Study," a large majority (57%) chose “improving responsiveness to demand fluctuations.”

To achieve this goal, even in uncertain times, CG companies must take essential steps to optimize demand responsiveness to maximize profitability. These steps start with increasing demand forecasting accuracy and layering on real-time data monitoring. Next steps include using advanced analytics to predict future demand and deploying artificial intelligence or machine learning to accelerate decision making and autonomous operations.

4 Steps to Supply Chain Resilience

Gone are the days of predictable markets for consumer goods products. Resilience is now the mantra for 2022 and beyond. Success depends on a CG company’s ability to make accurate plans and adjust on the fly. The goal is to strike a balance between having sufficient inventory to meet demand and freezing too many assets in a surplus.

“The variable costs in the supply chain from unplanned events are hard to manage,” according to the recently published RSR study, "Retail Supply Chain: Navigating Through Rough Waters with Improved Agility." This challenge jumped 17 points from the 2020, when it was chosen by 23% as a major problem to 40% in 2021.

The collapse of the well-oiled supply chain, which was built on many years of steady improvements, exposed CG-ers to the reality that no single strategy or method can successfully mitigate the recent string of disruptions that occurred during the global health crisis — manufacturing shutdowns, material and product shortages, workforce deficiencies, shipping snafus, and more.

Today, CG-ers understand change is inevitable and occurs faster than ever. Success depends on the ability to make rapid adjustments and pivot in real time. This in turn requires accurate data to use in sophisticated demand forecasts and execution plans that drive revenue and profits.

Step #1: Accurate Demand Planning Is the Starting Point for Effective Execution

Demand planning is the process of predicting demand for products to ensure they can be delivered to satisfy consumer need. A demand forecast is used to create a multi-faceted plan designed to maximize sales and profits while also keeping costs to a minimum. This is done by providing sufficient inventory to meet market demand while preventing margin-killing surplus.

When inventory is not aligned with the demand forecast, a cascade of problems occur in the supply chain. These challenges, according to the 2021 RIS "Supply Chain Technology Study," include too many out-of-stocks for fast-moving products (72%), too many out-of-stocks in stores (60%), too much inventory in slow-moving products (59%), lack of real-time data for fast reordering (58%), and too many out-of-stocks in the busiest locations (57%).

Accurate demand planning, based on accurate data, is the secret sauce that successful CG companies use to reduce the impact of crippling challenges that occur in uncertain times.

Demand planning begins with historical data plugged into statistical forecast models. These models become more accurate by including additional factors such as pricing, seasonal shifts, marketing, trade promotion management, economic trends, global events, competitive marketplace intelligence, and others.

Product lifecycle data points are also included to account for product interdependencies and the impact of the overall product mix on market share. Other important considerations include consumer trends, weather, corporate sales goals, and financial forecasts.

Once the statistical modeling process is complete, an initial forecast and execution plan is set in motion that guides supply chain functions and other departmental activities in the company, such as merchandising, marketing, and operations.

Step #2: Real-Time Data Monitoring Drives End-to-End Supply Chain Visibility

Performance monitoring is the next important step once the initial demand plan is set in motion. This is a critical step because it enables CG companies to obtain visibility into product movement throughout the supply chain.

For many CG-ers, two problems occur at this stage of demand plan execution — lack of end-to-end visibility and data collection that is not in real time.

Problems associated with data latency was cited in the CGT study titled “The Current State of the Evolving Global Supply Chain” where only 24% of CG-ers said they have real-time dashboard capabilities to give them end-to-end visibility.

CG-ers need end-to-end visibility in real-time to pinpoint inventory locations and status in transit. With it they can learn about delays and product deliveries schedules. They can track inventory quantities, predict demand flow, and make adjustments on the fly.

“Visibility into supply inefficiencies” was chosen as one of the biggest opportunities to address by 59% of respondents in the RSR supply chain study.

With real-time demand sensing and end-to-end visibility in place, CG companies can project day-two sales figures based on day-one trends and eliminate problems before they happen.

The more CG-ers know about what is happening in their supply chains and the quicker they know it, the more easily adjustments can be made to mitigate disruptions.

Step #3: Optimize Demand Forecasts and Plans Using Advanced Analytics

With end-to-end visibility and real-time data monitoring in place, CG-ers can use data mining tools and advanced analytics to make process modeling more agile to identify areas for improvement. Regular adjustments to the initial plan will reduce statistical errors and bias, help refine the forecast, uncover ways to cut operational costs, and boost revenues.

This level of demand planning requires predictive analytic tools that guide future supply chain operations. Predictive analytic forecasts identify the quantity of products to be purchased and sold during specific periods of time. However, it is not easy to achieve.

“Analytic capabilities that help measure the effect of supply chain activity” was chosen as a top inhibitor to efficiency by 59% in the RSR supply chain report.

To help solve this problem, the forecast starts with historical data available from internal resources. However, not all data is equally useful. It is important to use data from a relevant period. Data that is too old or from periods that do not correlate to current demands will produce inaccurate forecasts.

Also, it is important to use enough data, typically at least two years’ worth of historical data. If an anomaly has occurred in the previous two years, say a pandemic, it would be smart to include data from the previous three years. Using too little data, or the wrong kind of data, will reduce forecast accuracy.

Finally, it is advisable to use demand-based data rather than sales-based data. The difference is that sales data only shows the quantity of sales made during a specific period, while demand data shows how much sales could have been made if execution was optimized and out-of-stocks eliminated. Demand forecasts reveal true marketplace potential.

Advanced analytics today should include the ability to aggregate a robust range of data resources. This will include historical patterns, competitive marketplace data, consumer trend data, third-party data, POS data, social media data (sometimes unstructured), and more.

Today’s advanced analytic tools can also work with data anywhere — in the cloud, on multiple platforms, on-premises, hybrid cloud, and more. It should also protect data integrity, meet security and compliance standards, and run robust aggregations for complex analysis.

Step #4: AI and Machine Learning Accelerate Decision Making and Drive Autonomous Operations

Advances in ML and AI in the supply chain make it possible to adapt and update forecasts quicker than ever. Not only do they “speed up decision-making,” but just as importantly, “they pave the way for autonomous planning,” according to the McKinsey & Company report, "To Improve Your Supply Chain, Modernize Your Supply-Chain IT."

CG companies that use ML and AI can fine-tune their demand planning capabilities to allow inventory to run leaner without missing the mark on demand. In a survey of dozens of supply chain executives, 90% said they expect to overhaul their demand planning technology within the next five years, and during this transformation “four out of five expect AI and ML to be a key driver for supply chain IT implementations,” according to the McKinsey report.

New opportunities become available for CG-ers that deploy AI or ML in the supply chain, such as tapping them to create autonomous operations. The top three opportunities cited in the RSR supply chain report for implementing ML or AI include:

However, it is worth noting that CG-ers implementing AI or ML should take steps to link these initiatives to internal business values, standard execution processes, and departmental considerations, especially if used for autonomous operations.

At The Kellogg Co., for example, AI and ML learning initiatives are linked to a hard business case or specific marketplace problem that needs solving, according to Lesley Salmon, senior VP and CIO of Kellogg. “That ensures we are focusing our resources on something that will matter to our actual business performance,” says Salmon.

To make this happen, every function and business unit at Kellogg’s prioritizes use cases based on factors that include business impact, ability to advance the company’s technology, and ripple effect from change management.

Conclusion and Recommendations

By taking steps to make demand forecasting more accurate, CG-ers give themselves a wide range of opportunities that can be used to drive revenue and profits, especially an ability to make rapid adjustments and pivot in real time. 

By optimizing demand planning and execution, every item and location (distribution center, retail location, supplier, shipper, and so forth) is captured in granular detail that can be shared between trading partners. 

Future supply chain disruptions will never be eliminated; however their effects can be mitigated by implementing four key strategies:

These four strategies form a framework that creates a new level of supply chain resilience, which has emerged as the top priority in an age of disruption.