Development of an Improved Demand Planning Process - A Case Study at KåKå

Abstract

Background:
Operations planning and control has undergone an extensive shift from an individual company focus to a complete supply chain focus which has required a new common approach for supply chain planning and control to evolve. Increasing competition and globalization has created complexity in supply chain planning and integration since it requires a new cross-functional approach. To handle such complexity and difficulties, the cross-functional planning process Sales and Operations Planning (S&OP) can be used. However, to create a successful S&OP process the first step: demand planning that creates the work base for all S&OP activities, must be performed properly. To do so, forecasting activities must provide necessary input to the demand planning phase. This study aims to identify solutions to an improved demand planning phase in order to overcome the difficulties that supply chain planning and integration implies.

Purpose:
The purpose of this thesis is to improve the demand planning phase of KåKå’s S&OP process to create better conditions for more efficient operations.

Research Questions:
RQ1: How is the current demand planning process at KåKå designed and how does it perform?
RQ2: How can bakery products be categorized, based on characteristics, to simplify the forecasting and demand planning process?
RQ3: How can forecast methods be selected to fit different demand patterns?

Methodology:
A single case study with an abductive research approach, conducted by utilizing both quantitative and qualitative data.

Findings:
To summarize the findings of this thesis, the main identified issues of KåKå’s demand planning process are related to forecasting and product management. The use of one single forecast method for all products along with inadequate parameters, results in poor forecast accuracy. Also, the lack of clear processes, a large product assortment and product management being performed on an individual SKU level causes difficulties for both forecasting and demand planning. To face these issues, several potential solutions were identified in the analysis. Primarily, products can be categorized for forecasting based on their demand model to be able to allocate appropriate forecast method to each group, where the methods of simpler type resulted in the most robust and accurate forecasts. Secondarily, due to different grouping purposes, products can be categorized for demand planning by utilizing an ABC-XYZ analysis. The analysis should be based on two important demand planning characteristics in order to identify critical categories.

Contribution:
This thesis has been a complete elaboration between the two authors. Each author has been involved in every part of the process and contributed equally.

Popular Abstract

Demand planning is an important set of processes that enables companies to efficiently manage challenges with supply and demand. The act of demand planning is complex and it becomes increasingly difficult in the bakery industry due to seasonality and the sensitive nature of bakery ingredients. This thesis, on behalf of KåKå, develops an approach to master demand planning in the bakery industry.

The approach that has been developed to improve the demand planning process at K ̊aK ̊a requires a transformation of the current forecasting process. Primarily, products have to be categorized based on demand models to facilitate the choice of appropriate forecast methods. The demand model of a product can be evaluated based on three characteristics: demand variability, trend and seasonality. Secondly, forecast methods have to be selected to fit each product group to minimize the forecast error. One interesting finding in this thesis was that the simple forecast methods overperformed the complex and automated ones. By following this two-step approach and choosing simple models that are adapted to each product group, KåKå’s forecast accuracy can be significantly improved. This will address their problems with distrust in the forecasts despite a lot of manual work and make it possible to decrease the volumes of discarded products.

In addition to the transformation of the forecasting process, product management within demand planning has to be improved at KåKå. A second product categorization is necessary to avoid management on product-level. This was obtained by performing an ABC-XYZ analysis based on two important characteristics, which were chosen to be shelf life and sales value in this case. By performing this analysis, the most critical product groups can be identified. This categorization will also help facilitate the discussions and decision making in demand planning while ensuring that focus and resources are put where needed. To simplify the product management and keep the groups updated, it is also necessary to perform regular reviews of the assortment.

By implementing the recommendations, KåKå will be able to improve the performance of their current demand planning process. The recommendations are specifically designed to tackle the current problems that they are facing. The main problems have been identified to be due to the use of inadequate forecast methods and unstructured product management that, combined, result in high levels of discarded products. Both forecasting and demand planning were seen to be negatively impacted by the lack of structured processes, a large product assortment and the lack of product management on a higher level. One explicit result of these problems is the poor forecast accuracy achieved. This result should be positively impacted by implementing the recommendations presented in this thesis.

While the recommendations were developed to fit KåKå, it is believed that the approach can be relevant for other companies experiencing similar problems. The processes for product categorization and forecast method identification follow a general approach, while the applied parameters are considered to be industry-specific.

This popularised summary is derived from the master thesis: ”Development of an Improved Demand Planning Process - A Case Study at KåKå”, written by Emma Hall and Clara Kronberg (2023), Division of Engineering Logistics at The Faculty of Engineering – LTH, Lunds University

• author: Hall, Emma and Kronberg, Clara
• supervisor: Jan Olhager
• organization: Engineering Logistics
• course: MTTM05 20231
• year: 2023
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: Sales & Operations Planning, Demand Planning, Forecasting, Product Categorization, Non-manufacturing, Bakery Industry
• report number: 5991
• language: English
• id: 9123642
• date added to LUP: 2023-06-13 19:36:26
• date last changed: 2023-06-13 19:36:26