The role of the Master Scheduler has existed for years but organizations are still trying to determine the optimal way to manage supply. This has become more difficult over the years due to the volatility in demand, the increased pressure of balancing fulfillment, and the cost of capital. Organizations are striving to be demand-driven, but they require some degree of stability in the supply plan for supply assurance and for managing their capacity constraints― both internal and external. There are a number of factors that influence the master schedule. These may include:
a) The implementation of a postponement strategy or ‘delayed differentiation.’ Organizations that implement postponement strategies design a generic configurable product with customization closer to customer demand. The master schedule is generated at the generic product level.
b) Organizations that outsource. If you outsource you may be providing your partner a production plan which represents the plan at a higher level in the hierarchy. For example, product family. The partner will then be responsible for translating this to the item level master schedule.
c) Sales profile. If your product sales accelerate significantly at the end of the quarter (typically driven by sales incentives), a master schedule decoupled from the sales plan may be more important to you than other organizations. This is particularly important if you have long lead times or capacity constraints. Level loading inside a firm horizon will reduce the risk of supply to support the end of quarter sales.
d) Asset Utilization. Your organization may have significant investments in capital equipment that influence the master schedule. You may be faced with tradeoffs between capacity utilization and lean JIT strategies.
Once the master schedule is created, a big challenge is the monitoring of the master schedule against your production plan and your sales plan. How do you ensure that you are building the right amount of product to meet your organizational objectives? This can be a fine balancing act. You may want to ask yourself:
- Can you easily reconcile the production plan and master schedule at multiple levels of the hierarchy in units and value?
- Do you model your key bottleneck capacity constraints? Are they internal and external (ie. supplier) constraints?
- Do you have a clear view of your capacity constraints and their impact on your master schedule?
- Do you understand the impact your master schedule or changes to your schedule have on customer demand?
- Are you able to compare multiple versions of your schedule and analyze the impact on your KPIs?
- Do you use an alerting system to proactively identify when your master schedule is at risk of meeting your sales plan?
- Are you responsive enough to act before it is too late? What is your response management strategy?
We all understand the concept of creating a plan, but what about monitoring and responding to the plan? I am interested in hearing your point of view. Has the traditional role of the master scheduler changed? How important is this role in supporting an organization’s Sales and Operations Plan?
Discussions
Today, supply chain and its environment are not deterministic systems anymore. Supply chain is not only a dynamic system but also a stochastic dynamic system that its behaviour and state changes over time and most of the times we can not predict the future state of the system by certainty.
Planning for such a complex system is a very hard job.
there are two problems that have to be handled heres. First, we should try to develop compatible and harmonized supply chain plans in different levels. Second, we should handle the stochastic dynamic nature of our system in a way that make our plan look like robust.
Handling the first problem:
As we all know, three levels of planning are known as the hierarchical planning strategy which are strategic, tactical, and operational plans. the output of the higher level plan is the input to the lower level plan. For example, the output of the tactical plan (which is called aggregate production plan in manufacturing systems) is the input to the operational plan (which is called master production scheduling in manufacturing systems). The first problem that may arises is that while we are developing the higher level plan, lots of the details are not considered so probably when it comes to lower level plan, the higher level plan will show up not only problematic and imperfect but also infeasible! In order to handle this problem, some methodologies are offered by practitioners. One of them which is of my own interest, is establishing an iterative planning methodology that tries to develop the upper level and lower level plans iteratively up until theses two plans converge into a feasible integrated tactical-operational plan.
Handling the second issue:
Planners should try to be proactive to predictable events and disturbances and reactive to the unpredictable ones. Stochastic dynamic planning is of my own interest for this part. Demand and maybe many others parameters in the system such as lead-times, cycle times, ... are considered stochastic and stochastic forecasting methodologies are used to develop scenario trees. The outcome, is the tactical-operational supply chain plan that is the best even if any of the uncertain outcomes happen. So far, we have been just procative. What about reactive? Rolling horizon methodologies such as model predictive control concepts give us the upportunity to continuously track the supply chain and its environment and capture any change, unpredictable event and etc and send them as the new inputs into the planning module to decide about any possible change or modification we have to do about our plans. Butmaking changes are not that easy. System has to have the capability of accepting changes. It has to be fleaxible and also fast. Strictly speaking, it has to be agile.
In order to be agile, I personally like Dell manufacturing systems. They act in a kind of Push+Pull. From the upstream suppliers up to their assembly warehouses everything goes by push system and based on prediction (proactive). From assembly line to the final customers, they have a Pull system (reactive). It means that they don't assemble the product, up until they get an order for that. In this way, they can outsource everything and just keep final assembly processes as their internal job! So, their investment is in its lowest possible amount (internally).
I wish my comments help.
Regarding the first issue, I agree that there is a need to perfom volume/mix planning iteratively. Without the ability to translate volume to mix, gross assumptions are often used to relate volume to supply constraints which compromises integrity of the plan.
Regarding the second issue, I agree that organizations definitely need a process to be reactive. This is often referred to as response management. It involves the ability to quickly compare two sets of data inputs, understand the differences and quickly alert the user to the consequences of any change. The system has to be intelligent enough to consume, analyze and prioritize the data. The user can immediately start addressing the problem in a timely fashion.
Thanks
Leave a Reply