A More Resilient Supply Chain From Optimized Operations Planning Final
A More Resilient Supply Chain From Optimized Operations Planning Final
A More Resilient Supply Chain From Optimized Operations Planning Final
This article is a collaborative effort by Hossein Aghai-Khozani, Simon Bull, Valerio Dilda, Lapo Mori, and
Sebastian Reiter, representing views from McKinsey’s Metals & Mining and Operations Practices.
© Oatintro/Getty Images
September 2022
Across industries, companies that rely on day operations, understand true operational
transported materials for their operations have constraints, and use scenario analysis to ask
gained hard-earned knowledge from major critical “what if” questions.
disruptions over the past couple of decades,
most notably the financial crisis of 2008, As companies in manufacturing industries have
the COVID-19 pandemic, and geopolitical discovered, following these design practices
developments. During times of sudden upheaval, can increase supply chain throughput by 10 to
companies must quickly ramp down operations 15 percent in the short term, with no change in
and then ramp up again once demand bounces assets or overall configuration. Organizations
back, adapt their execution, and revisit long-term can also reduce costs by 5 to 10 percent and
plans. The supply chain has a special role to play, CO 2 emissions by 10 to 15 percent over the long
as companies rapidly shift their focus from cost- term while increasing operational flexibility and
effectiveness to maximized throughput—all with resilience toward disruptions. This article details
the same assets and infrastructure. the design best practices that support this effort
and how companies can get started integrating the
A poor response can have cascading effects, necessary capabilities into their business.
such as facility shutdowns or missed delivery
obligations. In recent years, approaches that use
data and analytics to identify answers and make Optimization in sales and
recommendations to specific business problems operations planning
have proved to be particularly relevant in bringing Optimization in operations planning involves
clarity to operations planning and thus improving determining the optimal choices for a set of
supply chain resilience. However, the vast decisions in a given business environment and
majority of companies have yet to implement such business target. This type of optimization generally
approaches in their supply chain, leaving a serious works best with prescriptive models that provide
gap in their planning capabilities. the ideal set of decisions as an output. The
elegance of optimization is its ability to adapt to
A recent McKinsey article1 examined three value not only changing business environments but also
chain approaches that can address this supply shifts in the business target—for example, from
chain gap—simulations of reality, optimization minimum cost to maximum throughput, highest
of plans, and real-time control-tower monitoring yield, zero environmental impact, or a combination
(see sidebar, “How analytics supports different of multiple factors.
planning approaches”). Here, we examine the
second approach in more detail, as it ranks among For these reasons, optimization is the ideal
the most powerful tools businesses can use to approach to readjust a company’s operations
navigate complex and changing environments, as outside factors or strategic priorities change.
especially disruptions. For example, an agricultural company recently
experienced dramatic rises in production costs
Organizations that want to get the most out of combined with limited transport capacity, creating
this powerful approach design their associated significant gaps in its ability to manage existing
optimization tools and processes along five best resource constraints. The company was able to
practices: improve information flows between respond by shifting its operations planning to an
teams, elevate customer centricity, bridge the optimization approach that almost completely
gap between long-term planning and day-to- closed this gap.
1
“Building value-chain resilience with AI,” McKinsey, November 26, 2021.
Supply chain analytics can support Simulations are descriptive analytics tools. resilience by expanding the scope of
planning efforts that broadly fall into three In a simulation, a digital twin is constructed real-time monitoring to anticipate and
types. While optimization has been at the with uncertainty measures and operational respond to upcoming or potential supply
center of this article, two other tools are inputs. Simulation is more appropriate for chain disruption. Real-time monitoring
simulation and monitoring. All three require comparing or modifying existing plans in the is the most useful when seamless
expert knowledge of the system, but face of uncertainty. Whereas optimization communication between various functions
simulation also relies on large historic data is more of a “black box” process, simulation and integrated decision making are critical.
sets (exhibit). produces explainable KPI-driven reports.
A typical simulation use case involves A combined approach is often best.
Optimization models are prescriptive examining an existing material network with For example, consider detailed railway
analytics tools. The main output is an uncertain production quantities, demand, scheduling. The complex rules and
optimal plan for the current environment. and transit times to identify first- and interactions require optimization methods
Optimization is most useful when an second-order bottlenecks. to create valid plans, while simulation
organization must create an ideal plan can then validate the performance and
from scratch that factors in complex rules The third analytics approach is real-time robustness of plans and determine the
and constraints. A typical use case for monitoring. Most companies have a most effective delay-mitigation policies.
optimization is building a monthly plan for control tower serving as an operating Real-time monitoring ensures seamless
operational production as part of the sales nerve center. However, fully digital communication between individual hubs in
and operations planning process. end-to-end control towers can increase case of unexpected disruptions.
Web <2022>
<Supply-chain>
Exhibit
Exhibit <2> of <2>
InIn
supply
supplychain
chainanalytics,
analytics,optimization,
optimization,simulation,
simulation,and
andmonitoring
monitoringcan
canwork
work
together but have important differences.
together but have important differences.
Analytics that can support supply chain planning efforts broadly fall into three types
Addressed Optimal execution plan for Identification and analysis of Real-time visibility of
business the current business root causes and bottlenecks for operations and financials
problem environment in alignment long-term resource allocation for decision support
with strategic direction
End delivery Productionized tool that can One-off analysis providing Continuously running
build and refine operational insights on bottlenecks data stream and
plans visualization
Technical Current live snapshot of Extensive historical data to Continuous data stream
requirement plant/process build relationships, of live operational and
distributions financial data
Insights and operational
information obtained Expert-provided insight into
through expert interview the decision environment
Web <2022>
<Supply-chain>
Exhibit
Exhibit <1> of <2>
Adopting
Adopting five
five best
best practices can help
practices can help improve
improve supply
supply chain
chain visibility
visibility and
and
coordination at all levels.
coordination at all levels.
Tool and process design must address all decision layers
Strategic orientation Operational execution
Supply chain
coordination
5 4 3 2 1
Best practices Customer Connecting long- Frictionless Understanding Scenario analysis
for optimization centricity term planning end-to-end true operational and critical “what
design with day-to-day information flows constraints if” questions
operations
Hossein Aghai-Khozani is a consultant in McKinsey’s Munich office, where Sebastian Reiter is a partner; Simon Bull is a
consultant in the Copenhagen office; Valerio Dilda is a senior partner in the Paris office; and Lapo Mori is a partner in the
Denver office.
The authors wish to thank Eddie Elizondo, Shailesh Lekhwani, and Tarusha Moonsamy for their contributions to this article.