In the course of doing a global study on the supply chain planning (SCP) market I talked to executives at solution providers from across the industry. One thing I asked them about was SCP in relation to the term digital twin. A digital twin is a digital replica of a physical entity or process. The term is most often applied to product lifecycle management solutions that help users design products and facilities. In asset management, these models take digital inputs (vibration, heat, etc.) from a machine, the inputs are analyzed, and are then used to more reliably and efficiently schedule maintenance for those machines. Increasingly the term is applied to supply chain planning. Supply chain planning systems are Big Data solutions that allow companies to hit their service level targets holding less inventory while using fewer human and physical resources.
But this sounded like a lot of hype to me. Some of the executives I talked to tentatively agreed with my viewpoint. Philip Vervloesem, a senior vice president at OMP, said “I like your reaction on this one. One could say that any action that place outside of pen and paper can be considered Digital.” (The figure of a supply chain planning model is courtesy of OMP).
But the great majority of executives working for companies providing SCP solutions really like the term digital twin, although they did tend to define its key features somewhat differently.
At the heart of a supply planning solution is a model of the supply chain process. The term digital twin may have a great deal of marketing puffery attached to it. But supply planning models are amazing. Using the term digital twin allows SCP executives to talk about SCP models, and how their companies’ approach to modeling is differentiated.
Omer Bakkalbasi, the chief innovation officer at Solvoyo, argued that a good digital twin for supply planning involves the ability to represent the supply chain process “at the right level of resolution…” What does Mr. Bakkalbasi mean by the right level of resolution? Supply chain planning is done for different forecast horizons – factory scheduling and fulfillment planning may be focused on what will be made and delivered in the next week. This is known as operational planning. Tactical planning is focused on what will be made and delivered in the month or few months. Strategic planning is focused on even longer time horizons – many months or even years.
“Resolution” refers to the level of granularity of a model. A model for detailed production planning requires a more granular model. These models have set-up times for each machine for each product, understand how much of a distinct product can be made on that machine in an hour, and understands production wheels (how setup times are affected by the order in which products are made). The tactical plans will be less granular, involving what Andrew Bell, the director of product marketing at Kinaxis, calls “approximations.” This model, for example, might have an average number of units (across many different products) that can be produced by the production line in an hour. Strategic models are even more abstract.
In theory, it would be great to use the detailed model at the operational level for tactical and strategic planning. But in practice, Hirish Iyer the vice president of industry & solutions marketing at Kinaxis, points out, detailed modeling “blows up run-times” (causes the computer to run for days or weeks before it can produce a plan).
In the supply chain planning market, a technology known as in-process memory allows for more granular models to be used but still provide plans in an acceptable timeframe. All relevant suppliers of SCP use this technology. Kinaxis, however, has developed their own patented, proprietary technology in this area, rather than using off the shelf technology solutions. The Kinaxis in-memory solution is focused specifically on using in-memory to speed up the speed with which complex supply planning models can be used and still provide plans in an acceptable time frame.
More granular plans are more “executable.” Mr. Iyer points out that supply chain executives engaged in tactical planning may question factory planners on why they did not execute the plan that was developed. But factory schedulers would reply that the tactical plan “never took into account how my machines actually operate.”
Concurrent planning is a feature that is increasingly in demand in the industry. It is the idea that though the operational and tactical plans may operate be based on different models, the plans should be interoperable. In a monthly planning meeting where executives are trying to match the demand they believe they will see for their products with what it is actually feasible to produce (the integrated business planning process), executives might come to the conclusion that overtime will be needed on a certain weekend to meet demand for a specific product. The executives may turn to a planner to run the operational model to see how realistic it is to produce the specified number of products over the weekend. With concurrent planning, the more specific model can produce that answer in just a few minutes because the two models are linked.
Kinaxis is best known for this capability. But in the course of updating my global supply chain planning market study, suppliers like Solvoyo, JDA Software, AspenTech, OMP, QAD Dynasys and others tell me they also offer this functionality. This means that the initial integrated business plan is more realistic; the plan that is being developed has a higher chance of being executed. It is a key planning feature that enables more responsive supply chain planning.
Currently, few suppliers have the capability to link strategic plans to operational and tactical plans. OMP and Solvoyo do talk about this capability, although when strategic and tactical plans are linked it is more focused on testing out new supply chain policies or seeing how changes to the facility infrastructure will change the company’s capabilities. In contrast, tactical to operational integration is more focused on the feasibility of a plan
Shaun Phillips, Director of Product Management at QAD Dynasys points out that it is the “common data model that ensures tight synchronization between operational and tactical plans.” Toby Brzoznowski, the chief strategy officer at LLamasoft, is also focused on the idea of a common data model. But he defines the digital twin somewhat differently than others in the industry. The digital twin is not the individual operational, tactical, or strategic model. It is a separate layer of the SCP solution stack that facilitates the strategic, operational, and tactical plans to operate from a common data – everything set-up times, to lead times, to bills of materials, to shelf space available for a product at an individual store. In short, the digital twin is the “end-to-end reference model always available” to operational, tactical, or strategic models. These different planning models than apply different statistical, machine learning, and AI algorithms, and different work flows, to answer different types of questions.
While I’m going to end the article here, this does not exhaust the topic. A good digital twin also gets data feeds that keep the supply chain model up to date. Further, the model is easy to configure and change. Finally, can we consider a demand planning model a digital twin? Those topics will be addressed in a subsequent article.
