Crossing the digital divide in optimizing supply chains. By Scott Zoldi

Optimizing supply chain strategies and network designs keeps getting harder. This task is stymied by persistent increasing challenges from customer demands, spiraling costs, disruptive competitors, and multifaceted partnership and logistical options.

These complexities are magnified by the maze of ERP, analytics, Excel, business intelligence (BI) and other tools required to manage the supply chain – everything from vehicle routing and fuel use to resource allocation and workforce schedules. When fast and accurate decisions are needed that impact the business’s performance, a supply chain can often break down in a sea of stalled IT requests, complex spreadsheets and myriad manual decisions. The complexity of working to solve this  supply chain challenge without a comprehensive analytics solution is both daunting and impractical.

Whether a manufacturer is looking to expand, drive a wedge between itself and its competitors, or simply balance growth and cost, analytics software can help, particularly where complexity makes ad hoc traditional analytics and BI ineffective. The problems that manufacturers face are getting increasingly challenging to solve given increasingly more constraints and variables than ever before. In some organizations, this is made worse by organizational inertia, in which they are hard-wired to do things ‘the old way’ – where advanced analytics solutions are questioned by domain experts and data scientists alike, slowing adoption and business benefit.

Increasingly, though, under time and financial pressure to cross the ‘digital divide,’ many manufacturers are leading in their spaces by building and deploying predictive analytic platforms that help their business experts solve mission-critical challenges with powerful, real-time, and easy-to-use optimization tools. These predictive analytic platforms enable analytic and business experts alike to leverage an optimization framework that taps rich simulation, visualization and reporting capabilities to transform business challenges into competitive differentiation.

For example, let’s look at how at how a leading global oil company addressed the challenge of creating a next-generation advanced process control platform that provides plant-wide control of all operations. They needed a solution that would maximize the economic operating benefits of each plant while minimizing overall cost, and used optimization software to solve complex problems with more certainty, reliability, and greater control. The technology was required to suitably plan for asset utilization and maintenance requirements while improving plant stability and profitability. The analytic software was used to advance operational efficiency utilizing characteristics of machines and processes while accounting for uncertainties in demand and availability.

Addressing these challenges with analytics software drives performance even with the most complex set of requirements. Optimization software provides powerful and versatile solvers that rapidly model complex 17processes with great accuracy, enabling a wide array of sophisticated, robust algorithms for automatically solving large-scale control problems.

This approach can deliver improvements that include:

  • Solving more complex problems. There are many interconnected units in supply chain operations that affect one another. The relationship between interacting variables, constraints and economic objectives leads to millions of decision possibilities.
  • Greater control and visibility. With optimization software, users are able to review a dashboard to see and respond to external disturbances,meet service-level agreements within the desired time frame, identify limiting factors and spare capacity, and optimize the use of raw materials while balancing the trade-offs with potential outcomes.
  • More reliability. Optimization solutions can deliver reliable, high-performance solvers, which enables confidence in making good decisions about the supply chain.
  • Rapid deployment. Rapid development of optimization models doesn’t deliver business value unless these models are rapidly deployed in production. Leading optimization platforms need to also reduce deployment times to close gaps between business insights/direction and operational improvement/value.

Another supply chain area where analytics software can bring great value is working to reduce production and transportation costs. These solutions not only deliver significant savings by maximizing payload, but they can also increase customer satisfaction by helping manufacturers consistently meet service level agreements.

For example, if a load reaches maximum weight but there is still space available in the trailer, then the payload has not been well managed and the company will need more loads to move the same quantity, resulting in more cost. When a load is close to the maximum cube and the maximum weight, it is considered to have optimal payload.

By applying machine learning, statistical models and prescriptive models to this area of supply chain operations, organizations can automatically recalibrate parameters and carry out ‘what-if’ analyses for scenario planning, allowing them to understand dependencies and sensitivities in decisions.

Predictive and prescriptive analytics software are completely changing the way manufacturers do business by dynamically and proactively managing all critical processes within their supply chain lifecycles. Leading analytic platforms are substantially reducing the time to develop and deploy optimized supply chain solutions, while also increasing visibility into every critical facet of the supply chain for the business and their constituents.

Scott Zoldi is Chief Analytics Officer at FICO responsible for the analytic development of FICO’s product and technology solutions, including the FICO™ Falcon® Fraud Manager product which protects about two thirds of the world’s payment card transactions from fraud. Scott is actively involved in the development of new analytic products utilizing Artificial Intelligence and Machine Learning technologies, many of which leverage new streaming artificial intelligence innovations such as adaptive analytics, collaborative profiling, deep learning, and self-learning models.