Latexco's well-considered choice for APS

Latexco is a Belgian family business and a global top player in innovative bedding solutions. Latexco's mission is “To help create a world where people can dream and enjoy a happy, active and healthy life”. 

In order to realize this mission, Latexco recently invested heavily in the further optimization of its production processes. In the past, production planning was based on coordination between employees of different divisions who structured their planning by collecting and processing data from often unintegrated, no longer adapted systems. 

Due to the increasing complexity of its own product range and customer requirements, this approach was no longer feasible. 

Latexco decided to invest in an Advanced Planning System, better known as APS application. 

To guide the company through this complex transition, Latexco partnered with Mentor and Men's consultants from the very beginning and to this day: from thorough business analysis, vendor selection, design of complex planning processes and custom roles, to change management. 

For more details on the elements underlying the decision for APS, we refer to the previous Client Story “Latexco - We support your dreams” . 

In this article we zoom in on the four basic domains of Planning that determine whether an APS solution is effective and suitable, and how the Mentor and Men-Latexco team dealt with each of these elements. 

1. Planning complexity 
2. Planning model & design
3. Planning organization 
4. Schedule dates 

1. Planning - Complexity 

The production planning and related processes could no longer be managed by adding 'people’s expertise and knowledge' to data from un-adapted systems. Hence, the initial choice for APS logic had above all a heuristic reasoning. 

Heuristics offer the advantage of being able to calculate a plan at high speed based on numerous business rules. Moreover, the results remain comprehensible for the human mind.  

For the majority of planning situations this approach proved to be the right solution. However, for the planning of very specific carousel machines in which more than fifty different molds – and thus different products – had to be able to be planned in a revolving carousel, these heuristics proved insufficient. For these situations we switched to a Linear Programming (LP). 

The next challenge was to configure the planning application for the operational logistics. This needed to be done as realistically as possible. Think, for example, about the transport of semi-finished mattresses from one machine to another, sometimes located in different geographical locations. 

To this day, this still requires adjustment. After all, you don't just take a mattress under your arm when it has to be moved to the next production step. 

With 245,000 variables, master data governance is vital. 

Latexco makes over 15,000 different end products that are partly produced on stock, partly on order and some of which have a Bill Of Material (BOM) structure of more than ten levels. 

2. Planning - Model & Design 

Apart from the carousel, the other production lines could each be modeled fairly unambiguously. The challenge was and still is to maintain an overview of the entire supply chain. 

It was Latexco's explicit wish not to work with so-called multistage orders. Latexco chose to plan each production step as a separate entity. This approach has the advantage that the output of each step can be registered as an intermediate stock and is therefore not simply known as WIP. The downside is that it significantly increases the amount of production orders to be planned and administered. 

We work according to the following planning design, controlled by the APS: 

Planning                    Frequency        Human intervention
Global SCP Model     Daily                 Validation by Production Planners
Detailed Scheduling  3xper day         Adjustment by Process Engineers

“This model seems obvious, but we noticed in reality - and in all honesty against our expectations - that one of the challenges was the reflection of the detailed scheduling decisions in the global SCP plan”, says Ruben De Gryse, Plant Manager Latex & foam at Latexco. 

Every Planner needed to have a correct and global overview of all production lines; a mandatory element to correctly estimate the impact of possible individual adjustments to the system’s planning suggestion. 

It was precisely in this mismatch between detailed scheduling and the global SCP plan that the biggest problem lay. It undermined the planning staff's confidence in the planning proposals. As a result, they regularly intervened on APS-proposals generated by the system. Proposals that were good at the base, but were nevertheless adapted or even rejected by planning staff due to a mistrust of the system. 

Fundamental system modifications provided the ultimate solution. A real feat, delivered by the team Latexco – Mentor and Men, and the application supplier. 

3. Planning - Organization 

Accepting that a system calculates for you was not easy for the planning staff. The systems calculates with 245,000 parameters (!) and proposes a production planning schedule that cannot always easily be understood by the human mind. In addition, the planning model is still being optimised, which requires extra flexibility from the planners. 

Also the employees on the production floor have to have more discipline to respect the planning sequence (scheduling) as proposed by the system. 

“We knew in advance that this important change was coming our way. Therefore, early in the implementation process, the role of production planners was carefully considered and a central 'production planning' cell was set up under the leadership of one Master Planner. 

People were trained and new ways of team collaboration were started. The consultants of Mentor and Men mentored and advised us with expertise, and with a lot of attention for a methodological Change Management approach”, says Liesbeth De Sutter, Planning & Customer Service Manager. 

Every Planner needed to have a correct and global overview of all production lines; a mandatory element to correctly estimate the impact of possible individual adjustments to the system’s planning suggestion. 

In order to increase the level of maturity in the planning, a demand planning process was designed in addition to the existing supply planning (deliberately implemented as first).  

The monthly preparation of a 'consensus forecast' in this demand planning ensured that this important activity had already been rehearsed and adjusted, and that the role of the demand planner was understood the moment the supply planning (SCP) started a year later. 

“The Mentor and Men team has gone to great lengths with us 
to thoroughly prepare and implement this transition”, 
says Lieven Vandendriessche, CEO of Latexco. 

Tight coordination between the demand planner (who checks, among other things, the safety stocks) and the master supply planner is necessary. It avoids that direct contact between sales and customer service departments on the one hand and production planners on the other hand leads to sub-optimal planning schedules. 

4. Planning - Parameters & data 

Data and parameters form the crucial basis for efficient planning. They had to be live and under control before we brought the planning systems live. 

In addition to the efforts to establish a strong planning organization, collecting and validating master data has been the most important and time-consuming activity, and has largely determined the success of the project. 

Here, we were strangely enough (or not) helped by the first Corona lockdown early 2020. During the quasi standstill in production, a team of fifteen Latexco planners and process engineers was able to fully focus on a gigantic exercise on data collection, including validation of production lot sizes and run rates. So-called intermediates were identified and defined. All corrected and newly created master data was migrated to that one master data repository in Microsoft Dynamics that was going to feed the planning system on a daily basis. 

The Master Data Manager, a newly created role, keeps the richly filled master database permanently up to date and ensures continuous optimization. 

With 245,000 variables, master data governance is vital. After all, any error in the master data can cause a planning run to fail or cause inconsistencies. “With our goal of Delivery Reliability in mind, we absolutely want to avoid such scenarios,” says Laurent Lepoutre, Master Data Manager. 

Focus on our promise of Continuous Delivery Reliability 

“During this entire APS process, we made a huge data leap. The quality of our data had to be improved in order to really optimize the planning. An absolute must to guarantee our ultimate goal of continuous delivery reliability to our customers”, says Jo Van Hoorde, IT Manager. 

“In addition to data and processes, the transition to APS planning required a major adjustment of the way of working of our employees. APS touches all divisions of our company. Everyone is involved. The Mentor and Men team has gone to great lengths with us to thoroughly prepare and implement this transition. A loyal partnership, based on mutual trust”, Lieven Vandendriessche, CEO.