Norwegian cities are becoming increasingly digital. Sensors, software and automation now sit in operational technology and buildings. Much of this is built and run by small and very small companies. Aiwell, where this PhD work was conducted, develops automated systems that solve practical problems related to snow, water and ice. When heavy rain, frost and meltwater meet ageing assets and complex contracts, reliability, security, and coordination become hard.
This thesis shows how small teams can use digital engineering to deliver safer, more resilient infrastructure. The core move is model based engineering: capture requirements, interfaces, architecture, threats and reliability as machine readable models linked to code, tests and operations. DevOps and DevSecOps connect those models to automated checks, so quality and security travel with the product from concept to installation.
The research has two strands. The analytical strand develops and applies formal models of attention, complexity and resilience in small teams. These models explain why failures arise and where small, well placed interventions pay off. The case study strand trials the approach in real projects with Aiwell and partners. Concrete outputs include lightweight patterns and checklists, game inspired workshops for threat modelling, and simple methods to set reliability expectations and test them continuously.
The payoff is fewer surprises, faster fault finding and evidence that reflects reality. Customers gain better inputs for procurement and operations. Small companies can demonstrate quality and security without drowning in bureaucracy. In short, digital engineering helps everyday infrastructure handle snow, water and ice more robustly.
I prosjektperioden har Aiwell digitalisert arbeidsprosessene og tatt i bruk enkle digitale verktøy i hverdagen. Vi dokumenterer nå krav, beslutninger og løsninger i felles digitale arbeidsflater med versjonskontroll og sporbarhet, slik at alle kan se hva som er gjort, av hvem og hvorfor. Faste maler og sjekklister gjør arbeidet raskere og jevnere, og vi gjenbruker det som fungerer på tvers av prosjekter. Dette har redusert feil og misforståelser mellom utvikling, salg og drift, gitt bedre overleveringer og gjort oppfølging enklere. Vi får tydelige varsler når noe krever handling, for eksempel når programvare må oppdateres, som bedrer sikkerheten uten å stjele tid. Nye medarbeidere kommer raskere i gang fordi de finner oppdatert informasjon på ett sted. Samlet sett leverer vi mer forutsigbart, bruker mindre tid på leting og omarbeid, og tar beslutninger på et tydelig grunnlag. Dette gir høyere kvalitet, lavere risiko og bedre samarbeid.
Den detaljerte sporbarheten har styrket kundestøtten. Når en kunde ringer, kan vi raskere se hvilken versjon som er installert hos dem, og basert på det vite om det finnes kjente feil som kan rettes med en nyere utgave. Dette sparer tid i felt og reduserer nedetid for kunden. Vi bygger også digitale modeller og bruker gjenbrukbare komponenter før vi går i produksjon. Det lar oss teste løsninger, sammenligne alternativer og oppdage problemer tidlig, før de blir dyre eller krevende å rette. Modellene og komponentene gjør det samtidig enklere å tilpasse systemer og dokumentasjon til kundens behov uten å starte på nytt. Vi kan gjenbruke byggesteiner, oppdatere dokumentasjon raskt og levere tilpasninger med færre feil
Model-Based Engineering (MBE) is increasingly being used in infrastructure projects, but so far these systems are expensive and complex, and not well suited to the small companies that make up majority of the infrastructure supply network. Leveraging systems modelling based on affordable and accessible digital modelling technology can help the small companies that make up the urban infrastructure sector deliver the reliable engineering required in complex projects. Small teams of engineers working in different areas can work on a common model, and significantly reduce the complications that arise in coordinating engineering activities. MBE can enhance the ability to capture, analyse, share, and manage information. Leveraging the potential for automation inherent in a model-based approach can support information re-use and make complex information easier to maintain. For small companies this can mean reduced cycle time, lower maintenance cost, and over-all improved product quality. The challenge is to determine not only the appropriate tools, but also the way to introduce these in the network of small companies involved in infrastructure projects. The study will use a mixed method grounded theory (MM-GT) exploratory sequential design for instrument development, and will be organized as Action Research (AR). The later phases will validate qualitative findings quantitatively with a larger sample drawn from industry clusters such as Smart Water Norway. The specific focus of this project will be Smart City water runoff infrastructure. Compared to other applications such as Smart Grid technology and building automation, this domain has been slower to adopt new technology. However, with extreme weather events and heavy rainfall increasing in intensity and frequency, there is an acute need to modernize how urban water infrastructure is designed, constructed and maintained.
