Dr. Evangelos Otto Simos is head of predictive intelligence at Kefallonia, Inc., a private research and consulting firm, and professor
at the University of New Hampshire. This report does not purport to be a complete description of global economic conditions
and financial markets. Neither the Journal nor Kefallonia, Inc. guarantee the accuracy of the projections, nor do they warrant in
any way that the use of information or data appearing herein will enhance operational or investment performance of individuals
or companies who use it. The views presented here are those of the author, and in no way represent the views, analysis, or
models of Kefallonia, Inc. or any organization that the author may be associated with. You can contact Evangelos at eosimos@
Dr. Nahavandi is Associate Professor of economics at Pfeiffer University School of Graduate Studies, specializing
in Business Economics, International Business, and Healthcare Economics. The information in these forecasts is gathered by the Journal from sources it considers reliable. Neither the Journal nor the individual institutions providing the data guarantee accuracy, nor do they claim that use of the data appearing herein will enhance the business or investment performance of companies or individuals who use them.
Many forecasters use Mean Absolute Percentage Error (MAPE) when evaluating the
effectiveness of their forecasting models. This paper discusses why MAPE may not, in fact, improve forecast model selection
because error metrics likes MAPE and RMSE often fail to pick up spikes which represent critical, demand altering events.
Here we detail the various limitations of this approach using theoretical and real-world examples, and present alternatives to
MAPE, which, depending on the situation, can provide much greater forecast accuracy.
Despite widespread migration to the cloud, a need for both centralized and decentralized
computing is emerging. Edge computing is a distributed computing architecture that brings computation and data storage
closer to the location where it is needed instead of in centralized data centers (i.e., the cloud), which improves response times
and saves bandwidth. Enterprises can leverage edge computing to maximize the value of the data derived from connected
things, generating insights much faster. Organizations can leverage edge AI to optimize a range of operations. At the same
time, Demand Planners can benefit by using it to generate faster data-driven insights, update forecasts in near real-time,
optimize inventory, and shape demand.
Here I present a case study of using S&OP and change management techniques to
turnaround the fortunes of a struggling plant. The plant was unprofitable and on time delivery was poor, and it suffered from
a series of systemic problems including siloed functions and a culture of blame that pervaded throughout the organization.
To tackle the underlying causes of this poor performance, we developed a solution based on 4 key elements: The Team,
Robust S&OP, System Health, and Root Cause Analysis. This article reveals how these elements were implemented and the
transformational change they brought about.
The typical approach to improving demand planning performance is to invest in expensive
systems, establish elaborate processes, and hire for specific functional skillsets. The performance of Demand Planners and
the overall function, however, depends on much more than this. A supportive culture is critical for any Demand Planner to
succeed, and key elements should be in place before we start analyzing Planners’ performance. Appropriate training, use
of the right tools, support from senior leadership, the right incentives, and non-quantitative performance metrics are all
valuable ways to set your team up for success.
The Covid-19 pandemic has exposed serious shortcomings in U.S. supply chains. Owing to
globalization and questionable supply chain management practices, the pandemic has revealed that the U.S is not selfsufficient
when it comes to food or medical supplies, presenting major economic and healthcare implications. This column
discusses learnings that can be gleaned from these two most critical of shortages and what can be done in future to fulfill
the most basic of human needs should a similar crisis occur in future. It also discusses the importance of Quick Response (QR)
initiatives in mitigating food shortages and how such initiatives are hindered by the fact that supply chains are fine-tuned
to be efficient instead of responsive.
Persistent forecast bias issues are unproductive and frustrating to those involved in
the demand planning process. Such issues are evidenced through excessive optimism and unreasonable overrides made
to otherwise realistic and fact-based projections. What makes forecast bias issues difficult to fix is that there is usually a
behavioral root cause associated with them. Considering this challenge, planning process practitioners can support their
organizations by designing and deploying mechanics that encourage and make the “right” behaviors easier to perform. In
this article we outline four levers that achieve this end: Use of Multiple Views, Opportunity for Collaboration, Assumption
Orientation, and Incorporation of Closed Loop. Once these levers are employed, we can expect forecast bias to decrease, and
forecast accuracy to improve.
Over the past 30 years, the approach to demand planning at most retailers and CPG
companies has remained stagnant, despite the increasing availability of data, advanced analytics, and technology. As a result,
retailers and CPG supply chains lack true demand visibility leading to manual overrides and increases in buffer stock. We can
use consumption-based forecasting and planning to solve these problems, gaining a clear picture of demand and linking it to
supply planning for the supply responses. This approach creates the much-needed ‘holistic supply chain’ and, in the coming
years, will become the new normal for retailers and CPG companies.
Dr. Nahavandi is Associate Professor of economics at Pfeiffer University School of Graduate Studies, specializing in Business Economics, International Business, and Healthcare Economics. The information in these forecasts is gathered by the Journal from sources it considers reliable. Neither the Journal nor the individual institutions providing the data guarantee accuracy, nor do they claim that use of the data appearing herein will enhance the business or investment performance of companies or individuals who use them.
During the current pandemic and subsequent demand disruption, we are preoccupied with understanding current consumer behavior and short-term demand. But equally as important is the longer-term picture. In this article I present a framework for understanding how demand for different products will shift, both as we exit this crisis and for the year ahead. I use a categorization system based on product needs, revealing how consumer behavior now and going forward will impact your demand forecasts.