Uncategorized – Traverse Consulting

Evaluating the Value of Advanced Planning and Concurrent Planning

Number of planners per value stream
1. The ideal is one planner per value stream for raw materials, intermediates, and finished products for the entire horizon.
2. Benchmark companies have no more than two planners per value stream for the entire horizon. The split in responsibility can be dictated by the nature of the supply chain, for example
  1. Split by time frame, short and long term
  2. Split by level of production, raw/pack and finished/intermediate.
  3. Split by production planning, distribution planning.
Ratio of Capacity to Demand
1. Low ratios of capacity to demand required that the production lines are always producing the highest priority products, as there is no spare capacity to waste, and limited recovery capability from any mismatches in demand and supply.
Number of products produced on each operation:
1. The number of possible schedules increases exponentially as the number of products per operation increases.
2. The planners task is to find the schedule that results in the lowest setup costs while maximizing customer service:
  1. 4 products = 24 possible production sequences
  2. 5 = 120
  3. 6 = 720
  4. 7 = 5,040
  5. 8 = 40,320
  6. 15 = 1.3 trillion possible production sequences
Number of possible production line options for each SKU.
1. Adds additional permutations to the number of possible schedules and increases the difficulty of finding the optimum schedule.
Efficiency losses due to setups.
1. An indicator that there may be value in finding the most efficient production sequence.
Number of levels of production.
1. The best schedule for one level is almost never optimum at other levels, and may not be feasible at the other levels.
2. The planner’s task is to find the best compromise that is feasible at all levels.
Efficiency losses due to intermediate inventory blocking or starvation.
1. An indicator that there may be value in better capability to model the storage constraints between levels.
Amount of storage between levels of production.
1. What is the amount of storage between levels, expressed as time? (# of days or # of hours at average production levels)
2. The chance of blocking and starvation increases as the storage time decreases.
Batch size and rounding coordination requirements between levels. For example, batch quantities of critical materials, pallets, tanks or containers, packaging materials, labels.
Are there intermediate operations shared by multiple value streams?
1. Visibility of requirements and collaboration is required to efficiently plan shared operations.
Non-production resources shared with other operations and degree of coordination required, for example staffing, tools, setup crews, washout stations.
Number of distribution centers.
1. As the number of distribution centers increases, demand variability at each distribution center increases, and safety inventory is divided among more locations.
Frequency of the planning cycle.
1. The length of the planning cycle is part of response time, decreasing response time decreases safety inventories.
2. Without better tools, increasing the frequency of the planning cycle will increase the planner’s workload.
3. Does the current system have a lower limit for the frequency of the planning cycle based on bucket sizes (daily/weekly), computing power, or computing time?
How often is the plan changed outside the normal planning cycle
1. Frequent changes to the plan outside the normal cycle consume planner effort, and are a symptom of a planning cycle that is too long for the current levels of inventory and variability.
Reaction time to respond to a plan being off track:
1. Sum of the time to recognize that the plan is off track, change the plan, produce, and ship.
2. Reducing the reaction time will decrease the amount of safety inventory required.
Number and nature of constraints.
1. An indicator that there may be value in a planning system that can accurately model the constraints.
Is coordination of production and shipping required at a lower level than daily time buckets?
1. A planning system that operates on a continuous time stream is needed in order to effectively coordinate activities at the daily level.

The Power of eLearning

When used properly, eLearning modules can drastically reduce the instructor effort required to deliver training, and simultaneously increase the student’s retention of the topic.

At P&G we were faced with a shortage of qualified instructors to support a global implementation of OMP+ Advanced Planning and Scheduling software. Our solution was to develop eLearning modules that allowed students to learn the basics of operating the software at their own pace.

Completion of all eLearning units was a requirement for attending the instructor led sessions.
The result was a 50% reduction in the instructor led training time and better retention of the material by the students.

The instructor led class could focus on the difficult topics that required interaction between the students and instructors. Student’s time became more efficient since they entered the class having already learned the basics of the software.

Additionally, this approach enabled offering remote training classes open to anyone anywhere in the world, further economizing on instructor time.

Re-Engineering the Planning Process for an Advanced Planning System

(Why you can’t plan the same way as you did before and expect improvement)

To gain productivity and create better plans with an Advanced Planning and Scheduling System (APS) you must do the following:

Precisely schedule the minimum horizon, roughly plan the remainder of the horizon.
Allow the system freedom to plan.
Focus on the real constraints.

Schedule the minimum horizon:

With less intelligent planning systems, planners typically spend time creating a long-term plan for capacity projections and supplier planning. An advanced planning system can create this long-term plan automatically, according to the rules and constraints you configure in the system. It adds no value to precisely schedule or manually adjust the plan in this region, as it will change anyway the next time you plan.

If you believe a good plan is not being created for the longer horizon, the solution is to revisit your constraints and planning rules, not to manually adjust the plan.

Allow the System the Freedom to Plan:

If everything is a constraint on the plan, it will be impossible to generate an automated plan that satisfies all constraints. The planner will manually adjust the plan to make it feasible, and then they will complain that the system saves them no time and effort.

If this is the case, let’s think about what they are really doing … they will be breaking one or more constraints.

To break this cycle, focus on the few real constraints that cannot be violated or broken, create weighted penalty costs that reflect other less important considerations, and allow the system to create the low-cost plan.

Focus on the Real Constraints:

Related to the point above, if the item is something that the planners will routinely violate in order to create a feasible plan, it’s not a real constraint.

The Business Case for Advanced Planning Systems (APS)

I’ve implemented over 75 sites for linear programming based Advanced Planning and Scheduling (APS) applications since 1989. An APS implementation is not to be taken lightly, as it typically requires re-engineering of the planning process and higher skill levels from the planners.

Here are some of the key indicators that an APS system will pay out:

Tightly coupled Make/Pack or Feeding/Consuming operations.
The difficulty of scheduling is beyond the scope of a single planner.
Constrained production.
Highly deployed inventory.
Re-scheduling is required within the day.
Coordination of production and shipping within the day.
- Otherwise, you could be planning to ship something at 10 am that will not be produced until 2:00 pm.
Complex spreadsheets are being used to plan production or distribution.