In spite of years of talk about how S&OP can be the process where strategy is translated into actionable operational plans, for many companies the process remains just a supply/demand balancing tool without much consideration of financial implications or supply/demand shaping. What’s needed, as Dean asserts, is new technology to enable strategic decision making for S&OP processes.
The traditional approach for linking strategy, aggregate financial planning, and S&OP has been bottom-up: companies first implement separate S&OP and financial planning systems, the latter often as an add-on to their ERP, and then try to extend these systems to somehow answer strategic questions.
This approach is fundamentally flawed. The planning tools associated with S&OP typically lack a sufficient horizon for strategic decision making. And while they are suitable for simulating a handful of alternatives – like evaluating various forecasting volumes in different markets, or different replenishment volumes given different inventory levels – they operate under the assumption that business processes and policies are in place, and the only task for technology is to implement analyses within this structure. Moreover, due to the “siloed” nature of these tools, the bottom-up approach can drive sequential decision making: decisions are locked in step-by-step, when what the company needs is to make simultaneous decisions across multiple steps.
These bottom-up tools have not been designed to do rapid prototyping of various scenarios or to come up with optimal solutions. They simply cannot readily answer strategic questions like
…and the list goes on.
These tools do not have the ability to model detailed constraints (business conditions) within or across demand, supply, and finance. Whatever high-level constraints are modeled are enforced using heuristics-based solutions (heuristics here refers to rules-of-thumb-based algorithms that are designed to do rapid and simplified planning). Initially, this approach may partially solve a problem and come up with a feasible solution. However, this “solution” will rapidly devolve into infeasibility once the solution becomes live and business conditions change. On the other hand, an optimization-based approach is able to quickly get a recomputed, feasible solution as long as the changes in business conditions (constraints) are communicated to the model.
In order to provide strategic decision making support for S&OP, what is required is a software solution built from the ground up, allowing companies to model the business in detail, and to proof test new approaches with minimal disruption. The key requirement of this solution is the ability to represent the demand, operations, and financial plans for any enterprise in an appropriate level of abstraction to develop optimized, rather than approximate, solutions to problems or questions.
It seems clear that neither S&OP nor aggregate financial planning, or really any conceivable adaptation of them that is based on summarization, can ever do optimization – at best, some logic based on heuristics applied to S&OP volumes might put you in the ballpark of a reasonable strategic alternative, but it will not provide an answer based on optimization logic.
All of this argues for a different type of software: one that operates outside the constraints of operational systems, that helps design business processes rather than enable them, that tries to evaluate various alternatives in some way other than trial and error – so that strategic alternatives are clarified for the decision-making process, and where strategic decisions can be translated into the supply-and-demand projections and financial plans in S&OP.
Here are two companies that have been able to perform strategic decision making in the context of S&OP.
A large energy-resources company was seeking to address looming shortfalls of conventional transport fuel by sustainable development of its region’s rich and well-located oil shale sources. The company implemented a prescriptive modeling, optimization, and analytics capability. The platform allowed the company to group the model’s key input parameters, such as the grade and moisture content of the mining output, and to create, analyze, and communicate to their mining and process teams a range of scenarios on the impact of variability in key inputs, including pricing variability in their finished products. The technology showed the company how to re-optimize plant design strategies and investment-timing options across the range of scenarios.
Cox Industries has been an innovation leader in wood product manufacturing for residential, utility, and construction markets. The organization invested in prescriptive analytics capability that identified opportunity value, defined as a fully constrained marginal value, for the next hour of production by their peeler machines. The result triggered a capital-evaluation project and ultimately the purchase of additional peelers. By utilizing prescriptive analytics technologies, this company bought two more peelers in geographic regions that maximized drying, freight, and sourcing constraints. Prescriptive analytics entirely changed the way Cox allocates capital and gave key decision makers a better understanding of the value that existing processes could create.
REFERENCE
Sorensen, D. (2016), Beyond S&OP and IBP to Enterprise Planning and Performance Management, Foresight, Issue 40 (Winter 2016), 2737.