The Data-Driven Decision Support Strategy Pt1 – Data Model

The collocation of data, processing into information and use of this information to drive accurate decisions support activities requires a strategic approach that cover 5 key areas.

  1. Data Modelling
  2. Data Input
  3. Data Attitude
  4. Data Quality
  5. Data Utilisation
5 Elements of Decision Support
The 5 Key Elements of a Decision Support System

Within each of 5 key areas there are several tactical areas the need to be in place to support an effective precision decision support process. I’ll start by covering the data model in this article. I’ll then move on to covered the other 4 areas in subsequent articles.

Data Model

“We are collecting the right data in the right systems to support the requirements of maintenance and reliability engineering analysis.”

Data Model Description
Data Model Description

The Equipment-Centric Data Model

The first aspect of the Data-Driven Decision Support Strategy is the data model. With maintenance and reliability of equipment (assets) the equipment must be at the centre of the model. The equipment function, i.e., the reason for its existence. What useful, value adding duty for the business is it required to do, e.g., pump oil, compress gas, detect fire, create water, etc.

Below is a diagram of the model. Its key purpose is to track the total cost and performance of equipment. This is required to answer the fundamental question of a maintenance organisation – is it optimising the balance between cost of carrying out maintenance vs. the impact of equipment functional failures.

Equipment Centric Data Model
The Equipment Centric Data Model

The Master Equipment List

An equipment record is required to capture the description, location and other technical information related to each item of equipment. Collectively this list of equipment is known as the master equipment list (MEL). It is a record of all the equipment the company uses in the transformation, transport, containment and control of the products it produces and the technical details of this equipment. This list usually stored in the CMMS database.

The Work Order

Next, we must log the maintenance work carried out against our equipment. This work is split into past work that has been completed and future planned work yet to be carried out. The type of work is broadly split between corrective work to restore the equipment function, i.e., after a breakdown and preventative activities required to prevent unplanned functional failures.

The work order is central to capturing this information. It stores the details of the maintenance planned and completed, including the type of maintenance, the cost of the maintenance and various dates including the scheduled start and actual completion dates.

The Work Order
The Work Order

The preventative and breakdown maintenance need to be separated. This is done using a code on the work order such as a work type. This is to allow reporting of both aspects of maintenance. It is vital to understand the cost of carrying out the preventative maintenance vs the total cost to the company of equipment failures. If the total cost of failures is lower than the preventative maintenance, then this doesn’t represent good business value. It would be more cost-effective to allow the equipment to run until it failed and then fix at that point.

Some CMMS systems have the concept of a work request and a work order. In this model the work request stores the details of the fault such as the equipment reference, the description of the defect, the equipment class, the failure mode, the component, the causes of the defect and a description of the work carried out to remedy the defect. The work order stores the planning and cost details, e.g., labour, vendor services, spare parts, special tools, cost centres and schedule dates, etc.

Equipment Performance (Availability)

The operational performance of equipment is the next element in the model. We need to know if the equipment was available for operation and how long it has run. If it is not available, we must know why. There are two possibilities: downtime for planned maintenance or downtime for corrective maintenance. 

Each of these need to be linked back to the work order created to carry out the preventative or corrective actions. This connection allows the company to build a picture of the equipment subject to the highest duration of downtime and, through the work order, the maintenance details of this downtime e.g. description of the downtime event, failure modes, failure causes, time to carry out the work, etc.

Daily Equipment Status Record
Daily Equipment Availability Status Record

Production Loss Tracking

It is vital that the consequence of equipment failures in production losses, environment and safety impact are tracked. These consequences may be logged in the CMMS but more often than not, are logged in other IT systems, each company will be different.

The key requirement for the equipment centric data model is that the consequences of these events are quantified and linked back to the failure of the item of equipment that caused the business impact through a connection to the work order. The benefit of linking through the work order instead of directly to the equipment is that the work order holds additional maintenance details of the specific event that caused the impact. It also stores the component that failed and the remedial activity that are carried out to remedy the defect. This includes the time and date of the event, the failure mode, the cause, etc.

Linking this consequence information to the item of equipment is a vital input int the performance analysis of the equipment.

Unplanned Loss Record

Next we will move on to discuss Data Input.

Share on facebook
Facebook
Share on google
Google+
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on pinterest
Pinterest

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}

Recent Posts


About the author 

Jason Davidson

Jason Davidson is the founder of EffectiveDashboards.com. He is a maintenance and reliability engineer who uses Power BI every day to create dashboards that help him and others make better decisions. In his spare time, he likes running and has recently got into photography. He also loves spending time hanging out and larking about with his two daughters.

FREE Power BI Quick Start Course

Learn how to build a Power BI dashboard in under 90 minutes.

>