Case Study

Production Planning Optimization:
Minimizing Set-ups

twitter
linkedin
facebook

The company

One of the leading players operating in the cork industry.

The challenge

Given the growth of the business volume, the planning activities were becoming increasingly complex and the inefficiencies too costly.

The capacity planning was done manually, considering typical constraints from the planning production formulation. Hence different possible machines for the same operation and a multistage production process were daily challenges for the person responsible for the production planning.

The execution planning was done by the shift supervisors. At the beginning of each shift, they would decide which orders would be produced; this approach led to an increase of set-ups, idle time and delays.

The approach

The first step towards meeting this challenge is to follow an integrated methodology, combining the main concepts of flow management by leveraging the KAIZEN™ methodology to build strong and efficient processes with the power of analytical skills.

All this starts with a clear definition of requirements, goals, and data collection/quality:

– List and question every production constraint – some were not relevant or were not a constraint.
– Set the goal of the algorithm to maximize output.
– Map every data flow and build network protocols to ensure data extraction and importation to the database.

Once these topics were addressed, it was time to move on to the production algorithms. In order to do so, it was important to set two different planning layers:

– Capacity Planning (medium level): Management of production capacity – determining which equipment and shifts are necessary to meet the proposed deadlines and objectives according to the demand variability.
– Execution Planning (low level): Sequencing of production orders, allocating them to a machine and a start time, respecting the sequence of operations and maximizing efficiency.

Note that these two layers are not independent: the execution layer cannot start its sequencing work if the necessary and adequate capacity to meet the demand is not present at the mid-level. So, it becomes vital to coordinate information from planning levels consistently and coherently.

Therefore, the first step was developing a tool that would plan the orders for a specific week and then a low-level tool that would minimize delays and set-ups on the shop floor. This two-step approach ensures a connection between the two planning layers, always supported by information connected directly to the systems.

Capacity planning

The capacity planning algorithm determines which equipment and shifts are necessary to meet the proposed deadlines and objectives according to the demand variability, allocating orders to groups of resources/machines.

An interface was developed to better visualize the results of the capacity planning algorithm that incorporated different departments of the factory.

Execution planning

The Execution Planning algorithm sequences production orders, allocating them to a machine and a start time, respecting the sequence of operations and maximizing efficiency.

This algorithm was run at the beginning of each shift in order to integrate the current stocks and WIP available.

The results were imported to a web app in which, with different logins, one could see different KPIs and add new shifts and machine parameters. A dynamic Gantt chart simultaneously allows a more extensive and detailed picture of the production plan. The user can quickly grab an order and try to change to another machine/sequence, and automatically a set of alerts appear in case the solution is not feasible.

Achievements

Considering the work done, the capacity planning was done more efficiently due to the increase of visibility of the different departments and the introduction of a set of rules to minimize set-ups.

The Execution Planning tool led to a better distribution of production time between machines and a reduction in the number of setups.

The main breakthrough of the project was the possibility to acknowledge when a certain order would be ready to send and the opportunity to anticipate an order and recalculate the production sequence.

#data analytics

See more on Analytics

Find out more about improving this business area

Get the latest news about Kaizen Institute