Simulating Ice Cream Production: Recognizing Constraints and Optimizing Production Plan

Problem:




Conaprole, the biggest dairy production company in Uruguay, produces more than 150 SKUs in their ice cream plant, using five production lines, and up to five different packaging configurations for each line.


The company plans ice cream production on a 12-month rolling basis as part of the Sales & Operations Planning process, and the demand plan varies a lot due to seasonality. The factory management needs to prepare the production lines for the peak season during the low season, taking into account product shelf life and the warehouse’s freezing cameras’ capacity and costs. The factory was often unable to meet the high season demand which generated stock-outs, and the management found it very difficult to quickly reschedule their detailed plans due to the challenges they faced.


Other factors, including bottlenecks and constraints in the production lines’ processes and variations in human resources’ availability that can occur randomly, made planning analyses even more difficult.


The management’s challenge was to be able to reformulate their plans in order to balance supply and demand and make sure they would avoid stock-outs in key products. They also sought ways to optimize the use of their production capacities. Ite Consult found simulation modeling to be the best tool to provide Conaprole with the solution to these problems.


The objectives of the created simulation model were:

  • To analyze various production scenarios for the following twelve months of changing demand. 
  •  To optimize production plans to avoid stock-outs for all SKUs. 
  •  To optimize the use of production lines’ capacity with the optimized production plan.

Solution:

Using AnyLogic’s discrete event modeling capabilities, the consultants designed and developed a solution integrated with the company’s S&OP planning platform and SAP Material Management and Production Planning. The created solution included three experiments with the model of the production system. Each of them addressed one of the objectives above and helped solve the business problems questioned.


In the first experiment, the model examined the initial production plan, detecting stock-outs and backorders that could be expected if production followed this plan. It allowed the management to explore production needs based on demand and initial inventory. This experiment also gave users the ability to find out, by manually modifying parameters, how different situations could impact performance, for instance: the need to close lines during certain periods, the necessity to modify equipment efficiency, extend resource availability, or change human resources’ schedules. Users could manually change priorities of SKUs and analyze the expected impact of such actions on revenue (costs associated with stock-outs’ differed by SKU). Additionally, they could define minimal used production capacity and some other policies.

Production Planning Model Logic

Production Line Simulation 

The parameter variation experiment ran the model of the system 100 times and searched for the solution to fulfill the demand and keep products’ shelf life as long as possible while minimizing warehouse costs.


The last experiment optimized the use of lines by freeing production capacity in peak periods. Production was scheduled as close to the beginning of planning periods as possible in order to leave free capacity in all production lines as a buffer.


The input data included:

  • Demand per SKU
  • Inventory levels per SKU 
  • Batch size, priorities, life shelf, warehouse space occupation 
  • Overall efficiency and mean absolute percentage error (MAPE), if needed 
  • Costs

The system considered the following:

  • SKUs’ allocation by lines and sublines
  • Lines’ and sublines’ capacities 
  • Line and packaging restrictions 
  • Warehouse limitations 
  • Production times 
  • Production schedules
Manufacturing Simulation Statistics

Model Statistics


Outcome:




All simulation results segmented by month and SKU were exported to Excel. Additionally, the model presented changes in demand and inventory levels by month in histograms. It also gave information about stock-outs, in case they happened.


By using the model, the Conaprole management was able to:

  • Discover the processes in each production line by SKU. 
  • Optimize production plans to better meet the demand while maximizing product shelf life and minimizing warehouse costs. 
  • Improve production line utilization to secure additional production capacity in case of an increased demand.

The simulation model provided the management with the insight to choose the solution that would increase revenue and minimize the risk of stock-outs.

More Case Studies

  • GE Manufacturing Plant Uses AnyLogic for Real Time Decision Support
    In 2012, GE opened a new battery manufacturing plant in conjunction with the launch of an innovative energy storage business. GE’s exciting opportunity brought on many new challenges, such as increasing production throughput and yield under evolving processes and uncertainties, and reducing manufacturing costs in order to gain market share. The GE Global Research Center sought out a powerful and flexible tool to analyze, not just the specific process, but the manufacturing system as a whole.
  • Shipyard Capacity Analysis
    With AnyLogic simulation software as the centerpiece, NASSCO utilizes a custom-built analysis system called the Large Scale Computer Simulation Modeling System for Shipbuilding (LSMSe) to provide highly detailed and accurate capacity analyses for both current production and potential new work.
  • Production Planning in Marine Industry
    The managers of one of the most important Italian yacht manufacturers needed a new, intelligent approach that would make the planning process simpler. The objective was to give the real production planner exceptionally rich planning information, which would allow the person to test and refine a plan before its implementation.
  • Simple Simulation Model Helps Intel Avoid Production Plant Downtime
    Intel factories used a particular type of equipment that often broke down, which caused capacity constraints. These expensive parts were used in critical factory operations, and the repairs took significant time, so it was necessary to have extra spare parts on hand to avoid downtimes. Broken parts caused constraints at some of the factories while other factories over purchased spares.
  • Improving Mining Outbound Logistics with Agent-Based Simulation Modeling
    One of the largest resource companies in the world, with over $80 billion in sales, decided to enter a new market. It was planning to build a new potash mine with 90% of the resources exported. They wanted to design a reliable supply chain, with a high speed of supply replenishing, and the ability to recover from natural disasters and man-made crises benefiting from such volatility. Amalgama and Goldratt companies contracted this project to design the potash mining operations and a full supply chain of outbound logistics.
  • Modeling and Optimization of Oil Production Using AnyLogic Fluid Library
    Canada is the third largest country to have oil reserves. However, most of the oil is in oil sand – the mixture of sand, oil, and water – which has to be heated up with steam to emit the oil. It is costly to maintain such a distribution system, and outages may lead to disruptions in steam injection and oil production. Stream Systems company applied AnyLogic simulation modeling to optimize expenditures and capture production lags
  • Highly Automated Production Line Planning and Optimization
    Centrotherm Photovoltaics AG is a global supplier of technology and equipment for the photovoltaics, semiconductor, and microelectronics industries. The company needed to identify the best automated production line and factory configuration to minimize costs and maximize throughput and reliability.
  • Simulation of the Construction of a Tunnel with a Tunnel Boring Machine
    The cost of one hour of down time of a tunnel boring machine is usually high and project managers have to do their best to avoid unnecessary delays in construction. The aim of the simulation project, which was carried out at Ruhr University Bochum in Germany, was to create a simulation model that would be capable of determining the bottlenecks in tunnel building processes in order to minimize the possible monetary losses.
  • Shipyard Proves Order Fulfillment Capability and Gains Visibility into Production Facility and Distribution
    Faced with a large order for diesel powered submarines, Admiralty Shipyards JSC must evaluate whether the current production facilities can fulfill the order, and if not, what amount can be produced by the year 2016. Admiralty Shipyards JSC is also seeking confirmation that an additional production facility will not be needed to complete the order.
  • Analysis of Management Strategies for the Aircraft Production Ramp-up
    The Airbus Group joined the European Union ARUM (Adaptive Production Management) project, which is focused on creating an IT solution for risk reduction, decision-making, and planning during new product ramp-ups. The project is aimed mainly at aircraft and shipbuilding industries. Simulation was chosen as a part of the ARUM solution, because it would allow the participants to reproduce the real production facility experience.